[House Hearing, 118 Congress]
[From the U.S. Government Publishing Office]
.
[H.A.S.C. No. 118-85]
FIELDING TECHNOLOGY AND
INNOVATION: INDUSTRY VIEWS ON
DEPARTMENT OF DEFENSE ACQUISITION
__________
COMMITTEE ON ARMED SERVICES
HOUSE OF REPRESENTATIVES
ONE HUNDRED EIGHTEENTH CONGRESS
SECOND SESSION
__________
HEARING HELD
SEPTEMBER 16, 2024
[GRAPHIC NOT AVAILABLE IN TIFF FORMAT]
__________
U.S. GOVERNMENT PUBLISHING OFFICE
57-661 WASHINGTON : 2026
-----------------------------------------------------------------------------------
COMMITTEE ON ARMED SERVICES
One Hundred Eighteenth Congress
MIKE ROGERS, Alabama, Chairman
JOE WILSON, South Carolina ADAM SMITH, Washington
MICHAEL R. TURNER, Ohio JOE COURTNEY, Connecticut
DOUG LAMBORN, Colorado JOHN GARAMENDI, California
ROBERT J. WITTMAN, Virginia, Vice DONALD NORCROSS, New Jersey
Chair RUBEN GALLEGO, Arizona
AUSTIN SCOTT, Georgia SETH MOULTON, Massachusetts
SAM GRAVES, Missouri SALUD O. CARBAJAL, California
ELISE M. STEFANIK, New York RO KHANNA, California
SCOTT DesJARLAIS, Tennessee WILLIAM R. KEATING, Massachusetts
TRENT KELLY, Mississippi ANDY KIM, New Jersey
MATT GAETZ, Florida CHRISSY HOULAHAN, Pennsylvania
DON BACON, Nebraska ELISSA SLOTKIN, Michigan
JIM BANKS, Indiana MIKIE SHERRILL, New Jersey
JACK BERGMAN, Michigan VERONICA ESCOBAR, Texas
MICHAEL WALTZ, Florida JARED F. GOLDEN, Maine
LISA C. McCLAIN, Michigan SARA JACOBS, California
RONNY JACKSON, Texas MARILYN STRICKLAND, Washington
PAT FALLON, Texas PATRICK RYAN, New York
CARLOS A. GIMENEZ, Florida JEFF JACKSON, North Carolina
NANCY MACE, South Carolina GABE VASQUEZ, New Mexico
BRAD FINSTAD, Minnesota CHRISTOPHER R. DELUZIO,
DALE W. STRONG, Alabama Pennsylvania
MORGAN LUTTRELL, Texas JILL N. TOKUDA, Hawaii
JENNIFER A. KIGGANS, Virginia DONALD G. DAVIS, North Carolina
NICK LaLOTA, New York JENNIFER L. McCLELLAN, Virginia
JAMES C. MOYLAN, Guam TERRI A. SEWELL, Alabama
MARK ALFORD, Missouri STEVEN HORSFORD, Nevada
CORY MILLS, Florida JIMMY PANETTA, California
RICHARD McCORMICK, Georgia MARC VEASEY, Texas
LANCE GOODEN, Texas
CLAY HIGGINS, Louisiana
Chris Vieson, Staff Director
Walker Barrett, Professional Staff Member
Phil MacNaughton, Professional Staff Member
Brooke Alred, Research Assistant
C O N T E N T S
----------
Page
STATEMENTS PRESENTED BY MEMBERS OF CONGRESS
Rogers, Hon. Mike, a Representative from Alabama, Chairman,
Committee on Armed Services.................................... 1
Smith, Hon. Adam, a Representative from Washington, Ranking
Member, Committee on Armed Services............................ 2
WITNESSES
Jenkins, Richard, Founder and CEO, SAILDRONE, Inc................ 10
Ludwig, Peter, Co-Founder and Chief Technology Officer, Applied
Intuition...................................................... 9
Sankar, Shyam, Chief Technology Officer and Executive Vice
President, Palantir............................................ 7
Tseng, Brandon, President, Shield AI............................. 5
Valentine, W. Mark, President and GM, Skydio Global Government... 3
APPENDIX
Prepared Statements:
Jenkins, Richard............................................. 116
Ludwig, Peter................................................ 103
Sankar, Shyam................................................ 87
Tseng, Brandon............................................... 76
Valentine, W. Mark........................................... 63
Documents Submitted for the Record:
[There were no Documents submitted.]
Witness Responses to Questions Asked During the Hearing:
[There were no Questions submitted during the hearing.]
Questions Submitted by Members Post Hearing:
[There were no Questions submitted after the hearing.]
FIELDING TECHNOLOGY AND INNOVATION INDUSTRY VIEWS ON DEPARTMENT OF
DEFENSE ACQUISITION
----------
House of Representatives,
Committee on Armed Services,
Washington, DC, Monday, September 16, 2024.
The subcommittee met, pursuant to notice, at 9:03 a.m. PST
at the UCSC Silicon Valley campus, 3175 Bowers Avenue, Santa
Clara, California, Hon. Mike Rogers (chairman of the committee)
presiding.
The Chairman. The committee will come to order.
OPENING STATEMENT OF HON. MIKE ROGERS, A REPRESENTATIVE FROM
ALABAMA, CHAIRMAN, COMMITTEE ON ARMED SERVICES
The Chairman. I want to thank the witnesses for being here,
and the University of California Santa Cruz for hosting us.
This is the first time in several years the Armed Services
Committee has left the bubble of D.C. for a hearing. We chose
to come to Silicon Valley to hear directly from America's
leading innovators about solutions to a problem that has vexed
the Department of Defense for decades: Why does the DoD
[Department of Defense] continue to struggle with rapidly
developing, scaling, and delivering innovation to our
warfighters?
For years we have been hearing complaints from industry
about the glacial pace of acquisitions and from small
innovators that lack the capital and support necessary to
bridge the valley of death. This is immensely frustrating to
us, because this committee has spearheaded dozens of efforts
over the last decade to reform DoD's acquisition processes,
create new flexible acquisition pathways, and make it easier
for the DoD to partner with private sector to expedite the
fielding of innovation.
It also is frustrating because our time to solve this
problem is running out. China has invested heavily in new
capabilities that are key to success on future battlefields,
and they figured out how to rapidly deploy them. Make no
mistake, these capabilities are being developed specifically to
defeat to defeat our military. We cannot let that happen. We
can't let China or any adversary outpace us on innovation.
Fortunately, the United States has something our
adversaries do not: a robust innovation ecosystem. There is no
shortage of innovative Americans, especially here in Silicon
Valley, with the ideas and know-how to keep us ahead of our
adversaries. We must take better advantage of this.
And we have seen in Ukraine the side effects--the side that
is faster at innovating, scaling, and deploying has the
advantage on the battlefield. Earlier this year we held a
similar hearing with senior DoD officials who insisted they had
all the authorities they needed and were doing everything they
can to expedite innovation. We are here today because we
respectfully disagree. We think more can be done and should be
done. We want to hear from American innovators on whether DoD's
acquisition pathways are actually working, and we are eager to
hear your recommendations for improving them.
Finally, we want to know your thoughts on how we can
overcome DoD's historic aversion to moving fast and taking risk
when it comes to innovation.
The Chairman. With that I yield to my friend, the ranking
member, for any opening statement he may have.
STATEMENT OF HON. RANKING MEMBER, ADAM SMITH, A REPRESENTATIVE
FROM WASHINGTON, COMMITTEE ON ARMED SERVICES
Mr. Smith. Well, thank you, Mr. Chairman, and mostly I want
to thank you very much for holding this hearing and making a
decision to come out here and have this incredibly important
conversation. I think it is, you know, one of the two most
important things that we need to do to get our national
security right, and to have the adequate defense that we need
for this country, personnel being the other side of that, which
I, you know, applaud your efforts on that, as well, in setting
up the Quality of Life Commission.
But I continue to be concerned that the Department of
Defense is not able to innovate quickly enough to keep pace
with our potential adversaries. And we have, as the chairman
has outlined, struggled with this for a number of years. We
have, primarily under the leadership of Mac Thornberry, when he
was chair of this committee, given the authority to the
Department of Defense to make innovations in a variety of
different ways. There is one important caveat to that, which I
will get to at the end of my statement, but that authority is
there, and yet we still move too slowly.
We are very focused on requirements and process, as opposed
to being focused on solutions. An example I have used many
times, I met with--Stanford has a thing called Hacking for
Defense, which is at a whole bunch of different universities,
where they take a group of undergraduates and give them a real-
world Department of Defense problem, something that DoD is
trying to solve, and ask them a question. Here is what we want.
And when I met with the students who went through that process,
every single group said the same thing: ``The first thing we
figured out is that they were asking the wrong question, okay,
that they were actually focused on solving something else. So
we pivoted and we adapted, and we solved that problem.''
Well, within the DoD world, that adaption and pivot is very
slow because the requirements, because the processes built in--
will spend years trying to trying to answer the wrong question
just because that is what was set in motion. We have got to be
able to pivot and adapt and move more quickly. And I think a
big part of it is culture within DoD. But all of you have
experience in working in that, and we would love to hear your
particular stories about what didn't work and, crucially, how
it could work better. We have a bunch of changes that need to
be made there, and we want to work on that.
The one caveat, yes, we have given the DoD a great deal of
authority, but at the end of the day we still appropriate.
There is the authority for other transactional authority
decisions and a whole series of other things where DoD can
theoretically make a decision to skip the normal requirements
process and move more quickly. But Congress lays out--they have
got to have money to do that. And if we appropriate down to the
last penny and restrict their ability to move it around, that
authority doesn't help them. So one thing I know this
committee--we need to work with our friends on the
Appropriations Committee to see how we can build in greater
flexibility so that the Pentagon can use the authority that we
have given them.
But I just want to close by emphasizing how important this
is. Whoever gets there first on the new technology has an
enormous advantage, and that has been true for as long as human
civilization has existed and tried to defend itself against
their adversaries. And there are all kinds of historical
examples of who figured out the machine gun first, or the tank
first, or the nuclear bomb first. Now this is happening weekly,
if not daily.
You know, new technologies are being developed for drones,
and for counter drones, for secure communications, or how to
disrupt communications so that your missile loses its signal in
mid-flight and can't hit its target. This is happening day in
and day out. We need to get ahead of that. And the chairman is
right, we should be able to do that. We are still the most
innovative economy in the world. Best universities, best
capital markets, entrepreneurship. We have got it, we just have
to figure out how to make sure the government is able to access
that in an effective way to give our war fighters what they
need to meet our national security needs.
Mr. Smith. And with that I look forward to the testimony
and, again, I thank the chairman for holding this hearing.
The Chairman. I thank the ranking member. Now I would like
to introduce our witnesses.
First we have Mr. Mark Valentine is the president of global
government business for Skydio.
Mr. Brandon Tseng is the co-founder and president of Shield
AI.
Mr. Shyam Sankar--did I get that right?
Mr. Sankar. Shyam.
The Chairman. Okay, Shyam Sankar--I knew I would mess it
up--is the chief technology officer for Palantir.
Mr. Peter Ludwig is the co-founder and chief technology
officer for Applied Intuition.
And Mr. Richard Jenkins, the co-founder of Saildrone.
I want to welcome our witnesses.
Mr. Valentine, you are up.
STATEMENT OF W. MARK VALENTINE, PRESIDENT OF SKYDIO GLOBAL
GOVERNMENT
Mr. Valentine. Thank you, Chairman Rogers, Ranking Member
Smith, and members of the committee. It is a real honor to be
here in front of you today. Thank you so much.
I am Mark Valentine. I am the president of our government
business at Skydio, where we are the largest drone
manufacturer, small drone manufacturer, in the United States
and the largest in the world outside of China.
Rapid acquisition is an important topic, and I am really
happy to be here today. It is also a timely one, because
acquiring the capabilities represented by the folks at this
table is absolutely essential for deterrence, and success if
that deterrence fails.
So my journey in this space began long before I came to
Skydio. As a combat fighter pilot I have a special appreciation
for the shift that we were seeing in military air power today,
and the way that small drones are actually transforming
warfare. The war in Ukraine is highly instructive in this next
generation of air power.
So when I was flying combat missions in the Middle East,
drones were relatively large, and they were very expensive.
Also, and even though they were providing outstanding support
to our troops and great situational awareness to our
commanders, they were very, very expensive, and typically
required lots of people to make them work. And, for the most
part, they were created to not work in a contested environment.
So the battlefield today, though, is fundamentally
different. Drones are smaller, they are smarter, they are
attritable, and often they are as important to ground personnel
as the rifles that they carry. In short, these drones are no
longer nice to haves; they are absolutely essential on the
modern battlefield. On the battlefield in Ukraine troops rarely
maneuver or even fire a shot unless they have some small ISR
[intelligence, surveillance, and reconnaissance] drone in the
air.
So we at Skydio actually understand this very well. We have
been sending drones to Ukraine since the beginning of the
conflict. We currently have over 1,000 systems in country now,
and we have many more that are on the way. Our employees have
gone to Ukraine more than 25 times in the last year-and-a-half,
and we have recently started hiring personnel to be there full-
time so that we can get access to not only assist end users,
but be able to make more rapid iteration based on these lessons
we were taking from the modern battlefield.
So these lessons that we are taking from also our
commercial drones are improving our products, which directly
help the United States military and our allies. As an example,
the AI [artificial intelligence] capabilities that are built
into our newest drone, the X10D, make it more survivable on the
electronic warfare regime in Ukraine and also in the urban
canyons of New York City. So based on these improvements, the
Ukrainians have requested thousands of the systems, and we are
getting them on the way as soon as we can.
These advanced AI and autonomy capabilities and our ability
to scale manufacturing are what is driving that success in
Ukraine and for our Department of Defense partners. We now
manufacture well over 1,000 drones per month, with the ability
to rapidly scale to over 2,000 very quickly. This has allowed
us to deliver thousands of drones on time and on budget to
every branch of the United States military, and support the
programs of record at the U.S. Army and the U.S. Marine Corps.
Our AI capabilities are also instrumental in DoD's AI for
Small Unit Maneuver program, which seeks to distribute the
intent of a single human to multiple autonomous systems. And
this is a game-changer for drones. From one operator operating
a single system to a single operator distributing their intent
to multiple robots, some that are in the air, some that are on
the water, and some that are on the ground, is the game-
changer.
So my written statement includes several recommendations
for improving DoD's acquisition of these capabilities, but I
would like to leave you with two things before we get to
questions.
First, the best way to improve U.S. drone capabilities is
to surge American-made small drones to Ukraine right now. Not
only is that going to help tilt the battle in the favor of
Ukraine, it is also going to give us the ability to take
advantage of an area where, bar none, is the most challenging
environment in the world. It is the best proving ground for
drones. So if we believe these drones are going to be valuable
in deterring conflict in the future, or if deterrence fails,
allowing us to prevail, then we absolutely need these systems
in Ukraine now so that we can iterate, learn, and improve our
own capabilities.
The second is that DoD's inventory of these small drones is
woefully inadequate for great power competition. Ukraine goes
through about 10,000 of these drones a month, and as of right
now our best estimate is that the entire U.S. Department of
Defense has about 5,000, and the procurement programs are only
procuring roughly 1,000 a year. So regardless of the
acquisition pathway we choose, the Department has to
dramatically increase these numbers. Failure to do so is going
to result in not enough drones when they are needed, and, more
importantly, a manufacturing base that is incapable of properly
scaling at the time of need.
And I don't share this as a drone company executive, I
share this as an American citizen and as a combat veteran and
fighter pilot who wants and absolutely needs our military to
have the capability to deter and, if that deterrence fails,
then to prevail in great power conflict.
So luckily, you all have the opportunity to rectify these
issues and accelerate the procurement of these drones and the
technologies at this table, and do so at a scale that makes our
adversaries think twice before provoking conflict.
So thank you so much for the opportunity to be here. I
really do look forward to your questions.
[The prepared statement of Mr. Valentine can be found in
the Appendix on page 63.]
The Chairman. Thank you, Mr. Valentine.
Mr. Tseng, you are up.
STATEMENT OF BRANDON TSENG, CO-FOUNDER, PRESIDENT, AND CHIEF
GROWTH OFFICER, SHIELD AI
Mr. Tseng. Chairman Rogers, Ranking Member Smith, and HASC
[House Armed Services Committee] members, thank you for hosting
this hearing on acquisition. The legislative work you are doing
is vital, and I appreciate your consideration and swift action.
And thank you for your service.
My name is Brandon Tseng. I am the co-founder and president
of Shield AI, a nine-year-old, multi-billion-dollar defense
technology company I founded with my brother in 2015. I am an
engineer, a former Navy SEAL, and a former surface warfare
officer with deployments to the Pacific, the Arabian Gulf, and
twice to Afghanistan.
Shield AI's mission and my mission is to protect service
members and civilians with artificial intelligence systems. And
to achieve this mission we are building the world's best AI
pilot, which is self-driving autonomy technology for aircraft.
Shield AI won the DARPA [Defense Advanced Research Projects
Agency] AlphaDogfight in 2020, beating all other AI pilots and
human pilots in simulation. We have more flight hours than any
company in the world flying jet aircraft autonomously. The
Secretary of the Air Force flew in one of our AI-piloted F-16
flights back in May.
We also build and manufacture an AI-piloted drone, the MQ-
35 V-BAT, which accomplishes the same mission as a $40 million
or $100 million aircraft at a fraction of the cost. Most
recently, this drone, the V-BAT, has been used by the Ukrainian
military to successfully execute a first-of-its-kind deep
penetration, long-endurance strategic targeting mission while
GPS [Global Positioning System] and communications were being
actively jammed, resulting in unprecedented effects for the
Ukrainians.
And what makes these operations possible is the cutting-
edge software that we build.
My perspectives on acquisition are informed by my
experiences working through acquisition problems every single
day at Shield AI for the past nine years, going from zero to
hundreds of millions in revenue, a very unique perspective few
have in this industry. And this experience leads me to some
recommendations for you today.
My first recommendation is shifting the DoD from a
requirements-based acquisition system to a problem-based
acquisition system. Henry Ford once stated, ``If I ask people
what they wanted, they would ask for faster horses.'' The DoD,
quite literally, has been buying faster horses for the past 60
years. This is because the DoD acquisition process is not built
to solve problems; it is built to fulfill requirements, which
takes anywhere from 3 to 20 years to validate and budget just
to get faster horses.
To remain relevant and competitive, the warfighter should
tell industry about their problems in the most intimate detail.
Then industry submits a novel solution. Don't give me a spec
that says, ``Thou Shalt Fly at 1,000 miles per hour for eight
hours.'' Instead, I prefer you to tell me, ``I need to have
strategic effects in this area, and this is what I am up
against,'' and ask me, ``How would you solve it?'' Then let
companies tell you how they would solve it.
For instance, Northrop may try to solve the problem with
the Global Hawk. General Atomics may try to solve the problem
with an MQ-9. Lockheed may offer an F-35. Shield AI may offer
an autonomous swarm of V-BATs to solve the problem. But the DoD
can choose the best solution for the problem, or a combination
of the solutions for the problem, instead of one that just
meets the requirements.
Second, the current House NDAA [National Defense
Authorization Act] contains section 833, Autonomous Unmanned
Aerial System Acquisition Pathways. This should be used much
more. This would create a contracting preference for the DoD to
use a dual acquisition pathway for integrating autonomy onto
new and existing military aircraft. The Air Force's NGAD [Next
Generation Air Dominance] family of systems program is already
using this process, and more programs should in the future. It
is worth noting that there is similar language in the Senate
bill, which will surely be conferenced.
My third recommendation centers around increasing the
funding and fielding of AI pilots, which is self-driving
autonomy technology for aircraft. This technology enables
unmanned systems to execute missions without GPS communications
or remote pilots. It also enables the concept of swarming,
which enables a single person to command hundreds of thousands
of drones effectively on the battlefield.
With Russia, China, and Iran jamming GPS and communication
links to stop our legacy drones and weapons, and their
proliferation of surface-to-air missile systems to stop our
manned fighter jets, AI pilots or self-driving autonomy has
become the single most important technology since stealth and
GPS-guided munitions.
AI pilots for drones and our weapon systems allow us to
restore air superiority, our most conventional--our most
strategic conventional deterrent. But today autonomy is funded
at levels that lack credibility or seriousness, largely because
requirements writers don't know how to write software
requirements. And without a requirement, there is no budget,
there is no program of record.
These are hard problems to solve, and I applaud you all in
Congress, the DoD, and industry leaders that you see seated
here before you, and the other companies present that are
committed to solving these problems.
Thank you for having me today on the committee.
[The prepared statement of Mr. Tseng can be found in the
Appendix on page 76.]
The Chairman. Mr. Sankar, you are recognized.
STATEMENT OF SHYAM SANKAR, CHIEF TECHNOLOGY OFFICER AND
EXECUTIVE VICE PRESIDENT, PALANTIR
Mr. Sankar. Thank you, Chairman Rogers, Ranking Member
Smith, and the distinguished members of the Committee, for the
opportunity.
I think we can all agree that we are in no ordinary state.
We have had more than 100 attacks on U.S. bases by Iran, 1,200
slaughtered in a pogrom in Israel, hundreds of thousands dead
in brutal combat in Ukraine, unprecedented tempo of gray zone
and phase zero operations by CCP [Chinese Communist Party]. It
is clear that we are living through a pretty hot cold war here.
And unlike World War II, where America was the best at mass
production, today our adversary is. And given the vast sums
that we have spent on defense in these decades of Pax
Americana, it would be reasonable to wonder what went wrong.
In 1993, after the end of the Cold War, America wanted a
peace dividend, and defense spending was slashed 67 percent.
The Secretary of Defense held a dinner in the Pentagon, the so-
called Last Supper, where he told the 51 prime contractors that
they were not all going to survive. Today there are five. The
actual consequence of the Last Supper is not a lack of
competition in the defense industrial base. It is actually the
fundamental decoupling of commercial innovation from defense
and the rise of the government monopsony. Consolidation bred
conformity and pushed out the crazy founders and innovative
engineers.
That is, until now. Today we are witnessing a First
Breakfast. America's commercial sector is re-industrializing
and innovating at an incredible pace. The challenge before us
is not whether America's industrial base is too small or too
slow, but rather why is the government unable to fully realize
the potential? And I think the antidote is simple: allow the
free market to build commercial solutions to problems that meet
the government's needs, and then actually buy those solutions.
Before the fall of the Berlin Wall, 86 percent of defense
spending went to companies that had both a commercial and
defense business. Chrysler made cars and missiles. Ford made
satellites until 1990. General Mills, the cereal company, made
artillery and inertial guidance systems. But today that 86
percent has become 6 percent. The monopsony's fixation on cost-
type contracting has made working in the national interest a
bad business, suitable only to risk-averse investors who are
addicted to dividends and buybacks. That is not what the most
dynamic parts of the American economy look like. That is what
the dying parts look like.
The last defense company to be added to the S&P 500 was 46
years ago until Palantir's inclusion this month. But we will
not be the last because today the founders are back, in the
hundreds, around this table with me, backed by hundreds of
billions of dollars of private capital to build in the national
interest.
So the question is, how do we harness this?
First industry needs to build, and that requires government
to buy. Ukraine expended 10 years of munitions in 10 weeks.
That is a clarion call that America needs to fire up
production. We need years' worth of weapons for our own needs.
Second, let us prove that the things that we have been
building with great treasure over the last decade will actually
even meet its moment in the modern battlefield. Ukraine shows
us that it is not what your weapons system is able to do today,
but how quickly you can adapt it to continue working tomorrow.
Third, if we want to compete with China we need to learn to
compete with ourselves more inside of government. We need more
crazy. The CCP is not going to know and be able to predict what
we are going to do, because we don't even know what we are
going to do. America's strengths are fundamentally creative and
improvisational, and that underscores the problem with
procurement. The predictability is a weakness. Highlighting
that rigid procurement process puts us at risk.
Everyone, the Russians, the Chinese have given up on
communism except for Cuba and the DoD. Five-year centralized
plans; a focus on costs, not value; measuring time spent, not
time saved.
If you want to start a new thing, you have to go ask for
money that you might possibly get two years from now. That is
actually insane. America's private sector has figured out how
to dynamically reprogram money inside of a quarter to get and
be able to buy what they need to win. The DoD ought to, as
well. And we must do that because we have to remember the only
requirement is winning. So a few actual recommendations for the
committee.
First, we should empower COCOMs [combatant commands] as
buyers. Even moving five percent of the budget to them enables
strategic competition with the services to ensure that we beat
back the worst instincts of the monopsony.
Second, while our system is actually quite excellent at
solving problems that can be solved deductively and top down,
it is horrible at solving problems that require induction and
iteration. Ironically, those things are the very strengths of
American culture, and that is why we need more Joint Urgent
Operational Need Statements, or JUONSs, or their service
cousins, the ONSs [Operational Need Statements], not less.
The folks who are crazy enough to submit JUONSs, they don't
toe the party line, people like Bill Perry, one of these
heretics. He pushed through stealth in GPS, not through PPBE
[Planning, Programming, Budgeting, and Evaluation], but by
going around it. Or folks like Admiral Rickover, who built the
nuclear Navy with 30 years of protection from Congress against
his own service. Heretics, but also heroes.
We at this table, those at the First Breakfast, we are not
just ready, we are painfully eager to ensure that America's
warfighters want for nothing. I don't think we require new
process or some massive overhaul. We just need the freedom to
do what industry does best: to build. Thank you.
[The prepared statement of Mr. Sankar can be found in the
Appendix on page 87.]
The Chairman. Thank you, Mr. Sankar.
Mr. Ludwig, you are recognized.
STATEMENT OF PETER LUDWIG, CO-FOUNDER AND CHIEF TECHNOLOGY
OFFICER, APPLIED INTUITION
Mr. Ludwig. Thank you, Chairman Rogers, Ranking Member
Smith, and distinguished members of the committee, for this
opportunity to testify at this important hearing.
As co-founder and chief technology officer of Applied
Intuition, it is a pleasure to welcome you to Silicon Valley.
We are grateful for your leadership and willingness to
highlight the important national security work in America's hub
of innovation.
Seven years ago, Qasar Younis and I founded Applied
Intuition with the goal of accelerating the adoption of safe
and intelligent machines. We began in the commercial automotive
industry, and now today work with 18 of the top 20 global non-
Chinese automakers. We produce some of the best software in the
world for unmanned systems, in both the commercial and defense
sectors. Specifically for defense customers, Applied provides
the software to rapidly field and safely update autonomous
systems. Our government-validated solutions are modular,
platform agnostic, and commercially proven.
Since 2021 Applied has worked with the Department of
Defense and was awarded three Small Business Innovation
Research contracts and a tactical funding increase this year.
These contracts highlight that the DoD recognizes the value of
cutting-edge commercial technology for military applications.
They also demonstrate growing interest by both the Army and Air
Force for applied software solutions.
While SBIRs [Small Business Innovation Research] are a
valuable entry point for start-ups, what matters most is how
quickly the DoD scales proven transformational technology. The
Defense Innovation Unit has played a critical role in aligning
commercial technology to programs of record. Through DIU's
[Defense Innovation Unit] Other Transaction Authority and the
Software Acquisition Pathway, the Army's Robotic Combat Vehicle
Program is using applied tools to test and evaluate autonomy
software. Unfortunately, the adoption of these agile
acquisition tools and firm-fixed-price contracts is lagging
across the DoD.
Meanwhile, China's commercial and military sectors are
poised to leapfrog the U.S. in autonomous systems, and there is
tremendous evidence for that in the automotive sector. In
response, the DoD must pursue thoughtful program design that
incorporates continuous development and integration. Our
warfighters need seamless software updates on the battlefield.
Software is never finished, and it becomes obsolete if it does
not evolve at the speed of relevance.
Second, the DoD should embrace ``buy before build,'' and
strong collaboration with the commercial industry. We applaud
the work of this Committee and the Appropriations Committee to
evaluate DIU and fully fund its innovation hedge fund. We
encourage Congress to provide continued political support and
robust funding for that unit.
Finally, the successful deployment of software capabilities
will require expanded use of the software acquisition pathway.
Program officers should be encouraged to more extensively
utilize these innovative pathways and firm-fixed-price
contracts. This is because they are outcomes-driven, scale
easily, and facilitate continuous improvements.
Applied Intuition is proud to accelerate the software-
defined force and provide warfighters the technology they need
to safeguard our national security.
Thank you again for the opportunity to testify, and I look
forward to your questions.
[The prepared statement of Mr. Ludwig can be found in the
Appendix on page 103.]
The Chairman. Thank you, Mr. Ludwig.
Mr. Jenkins, you are recognized.
STATEMENT OF RICHARD JENKINS, FOUNDER AND CHIEF EXECUTIVE
OFFICER, SAILDRONE, INCORPORATED
Mr. Jenkins. Chairman Rogers, Ranking Member Smith, and
members of the committee, thank you for the opportunity to
testify today. I am pleased to offer my thoughts on the DoD
acquisition process from the perspective of a small business
transitioning its products from demonstrated operational
success to large-scale, recurring fleet operations.
I am the founder and CEO [Chief Executive Officer] of
Saildrone, a U.S. company based in Alameda, California.
Saildrone is a world leader in providing oceanographic and
C5ISRT [command, control, computing, cyber, intelligence,
surveillance, reconnaissance and targeting] data solutions with
autonomous, uncrewed surface vehicles. Saildrone is one of the
first technologies to be funded by DIUx [Defense Innovation
Unit Experimental] and, more recently, DIU enabled a
significant deployment in 4th Fleet, supporting
counternarcotics and wide area domain awareness. Our unmanned
surface vehicles continue to rapidly evolve their capabilities
due to high cadence, year-round operations while facing real
adversaries.
In my view, only in the face of real, long-duration
deployments will solutions reach their full maturity and
utility.
Saildrones continue to prove their value, and have
undergone metric-based evaluations by OUSDI [Office of the
Under Secretary of Defense for Intelligence and Security]
(sic), amongst others. As an example, a recent report by the
Center for Naval Analysis on Saildrone effectiveness as a radar
platform reviewed four months of Saildrone operations of
continuous fleet operations. The report concluded the cost of
radar coverage to be around $4 per square mile from a Saildrone
and around $88 per square mile from a DDG [guided missile
destroyer], considering only operations cost, not platform
acquisition. That is a 95 percent cost reduction.
Now, I am not suggesting that a Saildrone can replace a
destroyer, but if we can use low-cost autonomous systems like
Saildrone to undertake roles that a ship would otherwise
perform, we can free up ship time for much higher-value tasks
and take advantage of their unique capabilities.
The same parallel exists for other Saildrone services such
as sea floor mapping and anti-submarine warfare, which can free
up time from the TAGOS fleet, destroyers, submarines, and even
P-8s, enabling them to be positioned elsewhere, doing high-
value tasks that only those assets can do. If you want more
ships at sea, then the fastest and most cost-effective way to
achieve this is to free up our existing fleet by removing long
endurance and persistent presence tasks that only unmanned
systems like Saildrone can fulfill.
However, despite proven performance, demonstrated cost
efficiencies, and multiple requests for Saildrone services from
different fleets, the service level budget process lacks the
flexibility to scale up these capabilities in the near term.
The current POM [Program Objectives Memorandum] process
requires funds to be requested multiple years in advance of the
money being spent. This creates a multi-year gap between--
[Audio malfunction.]
Mr. Jenkins. --deployed at scale, the classic valley of
death paradigm. To address this challenge, I ask you to
consider the creation of a dedicated bridge fund within a
service's budget, a fund to immediately roll out new
technologies that are proven effective, have adequate demand
signals, and are being included in the service's future POM
submission. DIU has greatly increased the Department's ability
to rapidly find and field commercial technologies. Now the
services need the ability to immediately scale their solutions
and not wait multiple years for dedicated bridge funds--
dedicated funds to arrive.
Critically, this bridge fund would not be intended for R&D
[research and development] or expanding funding for existing
programs. The bridge fund would instead specifically address
the valley of death years between validation technology and the
arrival of appropriations two years later. This approach would
solve two of the most significant pain points of the current
budgeting process.
Firstly, services would benefit from two years of
additional operational experience while the budgeting process
plays out. This technical de-risking is essential to enable
bold budget decisions which will be critical to modernizing the
future DoD.
Secondly, privately-funded companies would have headlights
for near-term growth, enabling additional capital investment,
scaling of inventory, growing manufacturing facilities, as well
as internally resourcing continued capability development.
[Audio malfunction.]
Mr. Jenkins. I was happy to see language in the fiscal year
2025 House Defense Appropriations bill, which provides DIU with
$240 million for finding innovative projects that the services
are committed to budgeting.
[Audio malfunction.]
[The prepared statement of Mr. Jenkins can be found in the
Appendix on page 116.]
The Chairman. I thank the witnesses. I now recognize myself
for questions.
Mr. Valentine, you talked about this large volume of drones
being consumed by Ukraine, yours and others. Tell me some of
the things that you have learned during this process of seeing
your product deployed and others in Ukraine.
Mr. Valentine. Yes, thank you, Chairman Rogers.
At a high level, I think you actually mentioned in your
opening remarks one of my biggest lessons, and that is speed
matters. We, I think, in the United States--I know from my
background in the military--we have faced the last 20, 30
years, where we have not been fighting a peer competitor. So in
our minds, we launch eight F-16s out, we always expect eight to
come back, and we need to get our head wrapped around the fact
that, in a great power competition or a great power conflict we
are going to launch those eight airplanes, and they are not all
going to come back. And what are we going to do?
And so what we have seen in Ukraine so far is, first,
drones actually are making a strategic difference. And if you
don't believe me, all you have to do is look to see how the
Russians have tried to counter to prevent the Ukrainians from
getting value out of that. And the way they have done that,
primarily, is through electronic warfare, jamming either the
radio frequency from the controller to the drone or, in many
cases, also jamming the satellite navigation systems so that
the Ukrainians can't get value out of those drones, they can't
get them to go where they need to go.
And they are so important to the Ukrainians because they
are using them as--back in my old language I would call them
forward air controllers to spot targets and then guide either
artillery rockets or human-powered FPV [first-person view]
drones to be strikers on those targets. So combating this
electronic warfare has been a rapidly evolving tete a tete,
make improvement, countermeasure happens, and it is literally a
cat-and-mouse game. And the only way to win it is to move fast.
And another part of this is that we are living in a
software-defined world, and once we realize we are living in
that software-defined world, the ability to rapidly iterate,
especially with companies that are dual-use, that are iterating
quickly in the commercial space that can now bring those rapid
innovations into the defense space and also take additional
lessons from there, that is a critical component. So moving
fast and then being able to do that at scale, I think, is the
biggest lesson and the biggest takeaway that I have had from
Ukraine.
The Chairman. At scale. Define what you mean, because you
talked about the volume that you have been putting into that
theater, and said that it was also other companies putting
large volumes. What do you consider a good scale?
Mr. Valentine. Well, I would look at the rate at which
Ukraine is losing small ISR drones, and that is roughly 10,000
a month. And I think that is probably a good benchmark. And
when I look at what they are losing per month and what we
currently have in our inventory, I think, my goodness, we would
last less than two months in a great power conflict. And I
just, personally, think that is unacceptable.
And I think there are some ways around it. Perhaps we start
to think about especially small, attritable systems like we
think of other attritable things like ammunition, and we start
to stockpile them. Whether that is in the form of finished
goods, whether that is in the form of at least the constituent
components being on our own shores so that if we do have a
demand shock we can rapidly assemble them, I think we have to
start exploring those types of things.
The Chairman. I was listening to Mr. Ludwig's statement,
though. If we stockpile this kind of technology, it is going to
be antiquated in six months. I mean, how do we get around that?
Mr. Valentine. Yes, sir. I think if you look at devices as
just a piece of hardware--it may or may not be true, but I go
back to the statement that we were living in a software-defined
world.
As an example, when we first sent drones to Ukraine, they
actually didn't perform very well because of the radio. But it
was a software-defined radio. And once we figured that out, we
were able to create a--take that radio, which was a multi-band
radio, and create a software-defined frequency-hopping schema
so that we could avoid the electronic warfare. And that actual
innovation, once we got that feedback, we made that software
change in days. We were able to test it in a few more days and
actually get it out to the field for field testing a few days
after that.
And so I think you can iterate and stockpile at the same
time.
The Chairman. Okay. Mr. Tseng, I would ask you the same
question. I just think Ukraine has been a great laboratory for
us to test things. What have you learned in your experience
with your products there?
Mr. Tseng. Well, 100 percent, sir. It has been a great
laboratory.
What I think the Ukrainians have discovered is that they
are not going to use anything that doesn't work on the
battlefield, period. And they--and the amount of U.S. equipment
that they do not use is staggering because it simply does not
work, and that is everything from our most exquisite weapon
systems to our--to cheap drones.
The Chairman. Give me an example. You can't just go --you
can't just throw that one out there and leave it.
[Laughter.]
Mr. Tseng. We have been on the battlefield in HIMARS [High
Mobility Artillery Rocket System] systems, right, our--which is
an incredible weapon system that I have used to great success
in Afghanistan. And look, there are variants of the munition
that work without GPS, right? These are the laser-guided
rounds.
But when you use a round that does not work or that is
principally reliant on GPS, the effects are highly volatile and
rare that you will actually hit things. And I think we need to
take a big look at that as we start to think about great power
competition, and have a really honest look at what we are
buying and fielding and saying, ``Will it work without GPS?
Will it work without communications?''
And I get nervous because I have been around, or been
involved with these U.S. military exercises that tout--and
there are very few of them--that we are going to jam GPS and
jam communications, and then they go back on their word. They
say, ``You know what? We are actually not going to do that
because too many of our things we know are going to fail,'' and
that is the complete wrong mentality. We need to see those
things fail in training in peace time so that we can prepare
for war.
The Chairman. Great. In your opening statement you said
that if the Defense Department--you said the Defense Department
needs to put serious money into this technology. What do you
consider serious money?
Mr. Tseng. I think it is--it would be on the strategic end
of serious money. That is how I think about it. And if you look
at our strategic capabilities and how much we are spending on
different--we spend billions of dollars on the adaptive engine,
we are spending billions of dollars on platforms. We spent $30
billion on the Manhattan Project, built the nuclear bomb, and
set the world order in four years. Right? This is the type of
technology that autonomy, that AI are, right?
Tesla has invested $10 billion into their self-driving.
Larry Ellison just last week said, ``Look, to write this new
age of large language model algorithms, you need $100 billion
to start.'' We are so far, as a Department, from any number
that is close to a billion, it pales. I would surmise no more
than $100 million is being spent on autonomy technologies writ
large in the DoD.
And so the number just needs to be higher, more
immediately, if you are really talking about getting after
strategic technology. And everybody in the DoD has said and
acknowledged this is a strategic technology. It is a strategic
capability. But if you go back to what I said originally,
right, all budget stems from the requirements. And if you don't
have people who can write autonomy requirements, then there is
not going to be any real budget behind it.
The Chairman. Great. Mr. Sankar, you made reference to the
structure. How would you structure the procurement process in
the DoD to be more agile and effective?
Mr. Sankar. Thank you. So when I think about my role, even
in my own company, the most important thing I am paying
attention to is where are we wrong. How quickly can I catch the
error in our roadmaps and our designs and fix that?
So how do we institutionalize and reward the sort of
defiance that we have from our warfighters on the front lines
of what is not working, and bring capital and resources and
innovation against those problems? That is where I think there
is a critical role for the COCOMs and their voice. They are at
the front lines. They are thinking through how they are going
to fight. They are going through the exercises, and they
understand and experience the gaps in the capabilities that are
being delivered by the services. How do we provide them the
power they need to go do the experimentation, build the
capabilities around that?
You know, the amount of innovation in this country, we are
unparalleled. The real question is, where have we bestowed
monopolies that prevent us from getting after that sort of
innovation, from confronting what are we actually wrong about?
You know, and I think the greatest missed opportunity with
Ukraine is that could have been our lend-lease moment. In the
counterfactual of World War II, without lend-lease it would
have been an axis victory here.
We have clear signal that we need to fire up production. We
also have clear signal that perhaps many of the things we have
built over the last decade will not meet their moment on the
modern battlefield. How many DMAG [Deputy Management Action
Group] issue papers have come out of what we have been learning
from the front lines of Ukraine? We have this sort of
generalized attitude of, well, those are two Soviet armies
fighting. We would fight differently. Things would be
different. All of that is obviously true. But the idea that
there wouldn't also be profound lessons to learn from those
front lines, I think, is not true.
The Chairman. You really got my attention when you made
reference to empowering the COCOMs in your opening statement,
and just referenced it again there. I really do believe that we
should be assigning each COCOM a pot of money that they have
got discretion over. They have to answer for it at the end of
the year, but they have got complete discretion over it.
I know I have a figure in my mind. What is the significant
figure in your mind that would be a difference-maker for the
COCOMs to have that kind of empowerment over?
Mr. Sankar. I think something like two to five percent of
the budget spread across the COCOMs would do a lot. You know,
my observation, having done this for 19 years now, is that it
is very hard to get a program and a service to wake up every
day and say, ``I got to fight and win against China.'' They got
their program, they got their cost schedule requirements, they
have got their--but it isn't that hard to get them to wake up
and say, ``I have got a PM [program manager] two doors down
from me that I need to wake up and kill.'' You know, the
competition within the government is vicious, and we should be
leveraging that for productive gains, the incremental delivery
of lethality.
So I think even just a small amount of budget reallocation
of the COCOMs introduces the necessary competitive signal
against all of our current other investments, what we are
spending the other 95 percent on, to actually see those things
go faster. It introduces these why should you adopt commercial
technologies that are cheaper and better when you are within
your cost schedule performance? Well, you are going to do that
if you think the good American two doors down from you might do
it first, and beat you, and make your program not as successful
as it could otherwise be.
The Chairman. Great, thank you. I will recognize the
ranking member for his questions.
Mr. Smith. Thank you very much. I am not sure about the
whole COCOM thing, but that is a whole other--they have their
interests in parochial things, as well, and I think we need to
make sure we set up the Pentagon so that the services are
focused in the way that you are talking about, but that is a
niche little conversation.
It seems to me that we are talking about two different
really big things here. One is process. How do we move off of
our requirements-based, stuck process that just, you know, gets
you locked into things forever, and move towards to the point
all of you have made, a problem-solving, flexible, adaptive
process that can change as it goes?
And then the second thing is, you know, even within that,
where do we spend the money? And that is a difficult, you know,
conversation when you are trying to anticipate where things are
going.
But it seems clear to me that we are spending too much
money on legacy systems from--well, and to Mr. Valentine's
point, in a non-contested world. We are imagining fighting the
way we have been fighting in a world that is not as contested
when we are moving into a different world.
And I would add to one thing you said there. It is not just
the great powers. I mean, the frickin Houthis are able to come
up with something. The barriers to entry here have become so
low that we are not going to go walking into Afghanistan,
right, anywhere in the world where we don't face an adversary
that has the capability of shooting down or sinking some of our
largest systems. So we need to pivot to that.
But focusing on the drone manufacturing issue, because it
seems like it has been a number of years now since it has
become clear we need to move in that direction, and yet we are
not manufacturing any significant numbers of the types of
drones that are so critical, so very specifically--and this is
for any of you--Mr. Valentine, Mr. Tseng, you have talked about
it the most, but what do we need to change to start increasing
that production capacity and building the number of drones that
we really need?
Mr. Valentine. Yes, thank you, Ranking Member Smith.
I think, in short, provide a demand signal. With a codified
demand signal, and knowing that there is a market there, then I
think not only Skydio but drone manufacturers all across the
United States can now start to purchase all the long lead items
that need to be --
Mr. Smith. What is stopping us from providing that demand
signal?
Mr. Valentine. Well, Ranking Member Smith, there are some
signals out there. But, you know, we hear words and rhetoric,
whether they are in the form of Replicator, this idea, that
idea. But quite frankly --
Mr. Smith. Well, let me ask you this --
Mr. Valentine. --I don't see much of that come to fruition
at this point.
Mr. Smith. To Mr. Roger's point--and there are reasons that
we don't--we give a lot of demand signals, okay, because they
are built in for years. But there is a central contradiction
between industry saying, you have to give us a demand signal so
we know what to build and, oh, by the way, you have to be
flexible and adaptive. All right? Because if we give you a
demand signal and then a year into it we are like, ah, we
learned something new, we don't want that anymore, now we are
locked into a contract forever.
This is the problem I have. I was very interested in ending
the monopoly that ULA [United Launch Alliance] had over Space
Launch. Okay, well, we gave ULA a really good demand signal,
and to a certain extent they produced, okay? And it was really
expensive, and then we got to the point where we needed to
adapt and innovate, and we couldn't because we had 10-year
contracts.
Mr. Valentine. Right.
Mr. Smith. So how do you balance those two things?
Mr. Tseng, you seem to have a comment.
Mr. Tseng. So, I mean, you asked the question, ``What is
stopping you?'' I think the requirements process is what is
stopping you.
The DoD, to points made earlier, has all the authorities to
go fast. So why can't they? Why aren't they? It is because all
of their money is already allocated and budgeted for something
that has a requirement --
Mr. Smith. Right.
Mr. Tseng. --which, if you have the requirement, then you
have fundamentally said you are not going to innovate anymore,
we are just going to fulfill this requirement and execute.
If you move to a problem-based system, then that money --
you are taking money away from the requirements process and
actually focused on --
Mr. Smith. So what would be the first step, then? Because
that is--I am very interested in that, and you--I think most of
you at this table have heard me talk about how I would like to
snap my fingers and eliminate half of the requirements, and I
don't care which half, just a good starting point. How do you--
how would we do that? How would we go in there and say, okay,
there is 5,000 pages of requirements. Those are in the garbage.
Solve this problem.
Mr. Tseng. I think I would--first I would mandate or I
would encourage the DoD that 25 percent of their acquisition
dollars in the next 3 years be spent on a problem-based
acquisition system. And from there, what you are going to see
is they are going to have to come up with an acquisition system
to--a problem-based acquisition system to actually hit that
target that you guys set out for them. And what that is going
to do, it is going to shift the flow of money from primarily a
requirements-based system. And over time we can get more to a
problem-based system.
And are there going to be hiccups, or there is going to be
challenges along the way? Yes. Do they hit 20 percent or 18
percent? Maybe. You know, but at least we have started that --
Mr. Smith. Which --
Mr. Tseng. --motion of getting them back in the right
direction.
Mr. Smith. --brings me back to the question that I had
asked --
Mr. Tseng. Yes.
Mr. Smith. --that we moved off of is, how do you balance
the need that industry is begging for, for a demand signal,
with the flexibility?
And Mr. Ludwig, you are shaking your head or nodding your
head there. So why don't you take a stab at it?
Mr. Ludwig. I think I really want to emphasize that the
importance of software and then, generally, agile methodologies
to all of these things--I strongly agree with the remarks from
Mr. Tseng about a problem-based system being highly
advantageous because, in our own work with the Department of
Defense, many times we feel somewhat restricted in terms of
what we can propose because of the requirements given to us.
Whereas, if the requirements are more so in the--in a problem
statement, we can actually provide a much more comprehensive
solution using more creative adaptations of our technologies.
Mr. Smith. Yes, it makes sense. It doesn't answer my
question, though.
Mr. Sankar?
Mr. Sankar. If I can offer--so I don't think you need 10-
year contracts. When people say, ``demand signal,'' I think
what it comes down to is what is the marginal time and effort
it will take to make a new fiscal purchase.
Mr. Smith. Right.
Mr. Sankar. So what is the fiscal OODA [observation,
orientation, decision, and action] loop? And, you know, private
capital will show up if I know, like, look, you can make a
buying decision every two months.
Mr. Smith. Right.
Mr. Sankar. I think even one year is too late and too slow.
Mr. Smith. Yes.
Mr. Sankar. So if we can get many more bites at the apple,
this is the fiscal version of DevSecOps [development, security
and operations].
Mr. Smith. Right. So to some extent, it is less a matter of
a demand signal and more a matter of a signal that we will
change and we will--you produce something we want, we are going
to buy it. Okay? And that is important, because, I will tell
you, the bigger primes, they like their demand signals, too.
And what they mean by demand signal is promise us that, no
matter what, for the next 10 years you are going to keep giving
us money, okay? And we have been locked into that in a very
crippling way. So I appreciate that distinction.
Last question. Where is DoD spending money right now that
we shouldn't be spending money? Because that is the second part
of this. It is a finite amount of money, okay? You want us to
spend all this money on drones? We got the budget we got, we
got a $34 trillion debt or deficit--either debt, actually. What
shouldn't we be spending money on that we are spending it on
right now, Mr. Jenkins?
Mr. Jenkins. So I thoroughly agree with the requirements
argument. One of my favorite sayings is a camel is a horse
designed by committee. I think LCS [littoral combat ship] is
well described by that. So yes, precise requirements or lack of
requirements and project-based solutions is key. You then have
a lot of people making new, innovative systems, but do they
actually work at scale that we need, right?
So the concept of a bridge fund approach was to give more
money for testing prior to end-of-lifing something. So the
process, zero net sum game, you know something is going to have
to go to get something new in. Those officers, those
individuals don't have the confidence in the new technology to
end-of-life something that is not proven. So running alongside
is the next piece.
To your question of what can you cut, I think I look at it
as how is the spend spread across the DoD. From my perspective,
which is at the lower end of the innovation loop, I see a lot
spent on R&D innovation. Now, I think the DoD spends $50
billion on innovation, and zero on go-to-market. As an example,
the Navy spends less than $40 million on fleet integration
projects a year. That is 0.01 percent of the R&D budget. If you
are a commercial company, a civilian company, you have a
product, you spend some money on R&D, you make a product, then
go-to-market strategy, customer testing, innovation testing,
supply chain, manufacturing, advertising, marketing, sales.
What we do is, as a DoD, as a country, is we spend all the
money on R&D and nothing on the go-to-market strategy. If you
were Apple, invented a new product but spent nothing on how to
make it, how to ship it, how to sell it, how to market it, and
just sat on your hands waiting for orders to come in, you would
have no sales.
Mr. Smith. I think that is a really good point that I had
not heard or thought of that way before.
Mr. Jenkins. So I think --
Mr. Smith. Yes.
Mr. Jenkins. --you need to equalize the money between the
R&D stage and then the testing and implementation phase before
the full-scale operations. Right now we are just missing that
--
Mr. Smith. Got it.
Mr. Jenkins. --which is the point of the bridge fund
concept.
Mr. Smith. I am about out of time here, but, Mr. Tseng, I
will give you the last word on that question of where we
potentially save money.
Mr. Tseng. Yes, sir. If the countermeasure to a system is
very cheap--the example being if a $1 million missile can blow
up a $400 million ship, or if a $1 million surface-to-air
missile can take down a $100 million fighter jet, then we
probably want to be buying less of those. And I am not saying
you are getting rid of every single one of them, right, but I
am talking about what the Air Force would call a high-low mix,
where we have a few exquisite systems that are augmented by
thousands, tens of thousands, hundreds of thousands, eventually
millions, of cheaper unmanned systems.
And so you just want to move that cost asymmetry advantage
to the United States versus where it lies right now, which is
with China.
Mr. Smith. So just to close, being provocative here, we
might not need 1,800 F-35s.
Mr. Tseng. I agree.
Mr. Smith. Okay, all right.
Thank you, Mr. Chairman. I yield back.
The Chairman. I like him.
[Laughter.]
The Chairman. I just want to remind everybody we are being
relaxed on the time, and we are going to remain that way, but
we also will have a second round. So if you think of something
after your time, we will come back around.
Mr. Gaetz of Florida is recognized.
Mr. Gaetz. Well, I think we are really getting somewhere
here. You know, you all have terrifically described the future
of warfare. And even a country lawyer can hear what you are
saying, that it has to be autonomous. And yet we are lashed to
a present where we are spending billions of dollars on stuff
that doesn't work, right?
It is not a bug of the system that we are not more rapidly
acquiring these technologies, it is a feature of the system. So
let's just--for quick math, President Biden requested 68 new F-
35s to purchase. Today in America, 29 percent of the F-35s are
fully operationally capable. Now, I don't know much about
warfare, like my colleagues, but I do know if something costs
$100 million, it should definitely work more than 29 percent of
the time, especially if you are telling us it is the past.
So we, I think responsibly, as the authorizers, we cut 10
off the block and said, ``You got to make more of these things
work.'' And then the Appropriations Committee not only restored
the 10 that we cut, they went and added 10 more.
So when you critique our system --
Voice. And didn't allow us to vote on it.
Mr. Gaetz. --Mr. Tseng--yes, and didn't allow a vote on the
good amendment--so that is what you are up against. What you
are up against is a corrupt system, where principally five
companies distribute hundreds of millions of dollars in
campaign donations and hire former staffers and hire former
lawmakers to be able to influence the process for them.
And I bet you didn't have me citing Elizabeth Warren on
your bingo card today, but she released a 2003 report entitled,
``Pentagon Alchemy: How Defense Officials Pass Through the
Revolving Door and Peddle Brass for Gold.'' And the key
findings of Senator Warren's report are that top defense
contractors hire hundreds of former government officials,
mostly as lobbyists. The defense industry consolidation
increases the risk for big corporations to abuse the powers of
the revolving door. And the big defense contractors hired the
most revolving door lobbyists and stocked their board with the
most former government officials.
So to all of you great innovators who are describing the
future of warfare to us, do you think it is a fair fight? Like,
do you actually think that if you come up with a better
mousetrap, that that is going to result in rapid acquisition?
Or is it just about who gives out the--let's see. Let's see. In
the last two decades, defense contractors have given out $285
million in campaign contributions and have spent $2.5 billion
in lobbying. So you guys think it is fair?
Anybody want to take that up?
Mr. Jenkins?
Mr. Jenkins. I would say no, it is not a fair playing field
by firepower. You know, small companies like Saildrone cannot
match the firepower of big, big primes.
I would like to see competition on the battlefield or the
pre-battlefield, and actually comparing technologies and
choosing a winner. As a, you know, commercial citizen, we are
very competitive, and we are happy to go head to head with any
other technology to prove that one is better than the other.
Mr. Gaetz. Yes. No, having a bake-off would be lovely, Mr.
Jenkins, but I don't think you are going to get that
opportunity, based on the rigged system we have.
And the fact that we are buying 10 more F-35s for billions
of dollars, that don't work--when Mr. Tseng says you guys are
spending less than $100 million on that, which is going to win
the future on AI and autonomous systems--is the fundamental
critique.
So the GAO [Government Accountability Office] does this big
report, and they say, look, we have got to have these lobbying
reforms where you can't have people rolling right into
influencing the very systems that they were a part of, and they
apply that to lobbying. And we, to our credit, passed that in
the 2018 NDAA. But then DoD started interpreting that to say,
well, that is just registered lobbying. That is not
acquisition, as well. So people engage in the revolving door on
acquisition, whereas on lobbying they are tightening down on
that.
Does anyone think that we should be more lax on the
acquisition revolving door reforms than we are on the lobbying
reforms? Does anyone think that?
Do any of you challenge the premise that the acquisition
process is corrupted when the senior Pentagon officials and the
senior generals involved in these programs then go work for the
big five companies? Do any of you say no, that is not corrupt?
Any of you?
Well, the silence is deafening. The silence is deafening,
because you all know you are playing a rigged game. And we
participate in it, and it is shameful. It should be the very
bake-off that Mr. Jenkins is describing.
But we will do all this stuff to learn about all these
exquisite technologies, but again, it is not a bug of the
system, it is a feature of the system, and it is deeply
unpatriotic.
A final question for you, Mr. Tseng. How much money--you
made mention of the amount of U.S. equipment in Ukraine that is
not being used on the battlefield because the Ukrainians don't
think it works. Do you have an assessment as to, like, how much
money that is?
Mr. Tseng. It is in the billions.
Mr. Gaetz. The billions. In the billions. And so U.S.
taxpayers are paying for the inflation to send stuff to Ukraine
that doesn't even work, and it is magnitudes more than on the
stuff we are spending for ourselves that does work.
Thank you for your testimony. It is illuminating and
enraging all at the same time.
I yield back.
The Chairman. I thank the gentleman. The chair now
recognizes Mr. Khanna of California.
Mr. Khanna. Thank you, Mr. Chair. Thank you for having this
hearing in Silicon Valley to you and to the ranking member. And
we are in my district, so welcome to both of you and to every
member here. It is fantastic.
I don't know, with Gaetz quoting Elizabeth Warren, I am
concerned he may be gearing up to run against me here.
[Laughter.]
Mr. Khanna. I hear his wife likes it out in Half Moon Bay.
So it is a great district that I hope everyone will get to see.
I was struck, Mr. Tseng, by your statement an adversary
military with a budget of 25 billion that effectively employs
lower-cost drones and autonomy will be able to decimate a
military with an 800 billion budget without drones and
autonomy, and I guess I am trying to understand the balance.
You know, I actually put into ChatGPT--not that it is
authoritative--what would be the top seven things to fight in
China, and swarm drones came out sixth. I am not saying it is
correct, but, you know, F-35s and aircraft carriers and
submarines still matter. I mean, they--it is not like we can
just have a military of drones. And yet we need to have some of
the new technology to be competitive.
And I guess the two questions I would start to ask is,
first, we have known on the F-35s, as Mr. Gaetz and the
chairman and others have said, that there have been cost
overruns, that they haven't been effective. Do you think any of
the technology companies here providing more contracts or
competition will do anything to improve the F-35 delivery, or
is there--what do we need to do to improve the traditional
platform delivery?
Mr. Tseng. I think the companies here could, in concert,
make the F-35 more effective on the battlefield in terms of its
employment, but I couldn't speak to how you could increase that
29 percent fulfillment rate. I don't think that is our
company's specialties here.
Mr. Khanna. And there is nothing that any of the new
technology companies could do eventually to compete with the
primes in terms of--you know, when you say ``problem solving,''
my guess is DoD will come back and say, okay, to solve the
problem we still need a lot of legacy traditional platforms.
So are you saying here that your value add would be the 20
percent where they say, yes, we need swarm drones and other
things? Or could you also in any way compete on the 80 percent?
Shyam?
Mr. Sankar. You know, 50 percent of our business is
commercial. We help people build planes, trains, automobiles,
ships. The largest ship manufacturer in Korea runs on our
software. We help GE [General Electric] build the J85 jet
engine that goes into the T-38 trainer. We helped Anduril build
their latest weapon systems. So I think there is a huge amount
that Silicon Valley and technologists can do to change
production of existing systems here.
I think to everything Mr. Gaetz said, I would agree, but I
would also say one of the deeper issues is the dysfunction and
pathology of the monopsony. You know, having a buyer that has
kind of unilateral control, it doesn't get to benefit from any
of the market forces and innovation. You know, in the
commercial world you don't have to be the smartest company; the
aggregate signal from all of the companies allows you to
innovate and capture things. And so when you have a buyer who
has to ``figure it out all on their own,'' and then is
convinced they can't be wrong, it is quite disastrous.
So I think many--if you think about the F-35, I would love
to help with the readiness the same way that every--I manage
over 65 percent of the world's air fleet, commercial air fleet,
but I don't have access to managing the DoD's fleet or helping
out in any way, shape, or form. These things are locked up
vertically within the prime contractors in their programs.
There are different rice bowls here.
So I think there is a huge sea change that we could have --
Mr. Smith. And I am sorry, Mr. Chair, important to this
point--we could maybe pause Mr. Khanna's time, but there is a
way to go after the larger systems.
The B-21 was built much differently--is being built much
differently than the F-35. And part of it was it was done with
other transactional authority, so it skipped a lot of the
requirements process. And crucially, it maintained competition
within sub-systems, instead of like what we did with the F-35,
where we just gave it to them at the start of the process. So
they owned us at that point. Vendor lock.
So yes is the answer to your question. Whether you are
talking about building an aircraft carrier or a drone, you can
do it in a way that maintains competition and gets out of the
requirement-based process, which is crucial, to your point,
because we are going to have to spend money on those things.
Thank you, sorry.
Mr. Khanna. And I would just say--and then I will give you
the last word, Mr. Ludwig--I think it is important for the tech
companies, yes, talk about drones, AIs, but also talk, in my
view, about how you are going to make the traditional planes
and weapons more competitive, because I don't think you are
ever going to convince the American people that we could just
have an American military of drones and AI in the next 10
years.
Mr. Ludwig?
Mr. Ludwig. To Mr. Smith's point on competitive sub-systems
and to Mr. Khanna's point on F-35 competitiveness, I do want to
emphasize, especially, the capabilities of these companies in
supporting extremely advanced software on these systems, right?
The next generation of these vehicles in the concept of
autonomy, that is mostly software. And in the case of the F-35,
that is not a software company, right? The primes are not
software companies. And the way that their organizations are
structured view software as a cost center, not a profit center.
And that means they are not getting the best people from
industry in those software teams.
The companies here on the panel, we are largely software
companies, and we are fighting for the best talent in the
market, and we can provide that talent to work on these
systems--as an example, adding autonomy capabilities for the F-
35.
Mr. Khanna. Thank you, Mr. Chair.
The Chairman. I thank the gentleman. I want to remind
people I am not being the usual Mike Rogers who enforces the
five minutes.
[Laughter.]
Mr. Gaetz. We are so well trained.
[Laughter.]
The Chairman. That is right.
Mr. Gimenez is recognized.
Mr. Gimenez. Thank you very much. You know, Hyman Rickover,
the father of the nuclear submarine force, he was rewarded with
his vision by being reassigned and also given an office in a
ladies bathroom.
Billy Mitchell, he was rewarded for his vision for air
power and the dominance of air power in the future, air
warfare, by being court martialed.
So, you know, we are--I am wondering if the same kind of
mentality still reigns at the DoD, where anybody who dares to
challenge the orthodoxy is put away. And so, Mr. Tseng, you
talked about your AI-powered F-16s and they did very well. How
did they measure up to the piloted F-16s in simulated combat?
Mr. Tseng. Oh, win 99.9 percent of the time.
Mr. Gimenez. Which is the worst it will ever be.
Mr. Tseng. So --
Mr. Gimenez. Right, because AI is still in its infancy.
Mr. Tseng. Yes.
Mr. Gimenez. So it is just going to get better and better
and better. So do you believe, like, that it is possible that
the F-35 could actually be the last manned aircraft, fighter?
Mr. Tseng. I believe it should be, personally. And I think
you can augment it with swarms of unmanned fighter drones,
swarms of lower-cost drones, yes.
Mr. Gimenez. And I disagree with the ranking member. I
think the American people would welcome our machines fighting
for us, so we don't lose our young men and women in battle. I
am ready to go there right now, okay, if --
Mr. Tseng. Yes.
Mr. Gimenez. --we actually win the battle.
Mr. Tseng. Yes.
Mr. Gimenez. Okay, so --
Mr. Smith. Just to clarify here, I was talking about how
the systems were built, not about whether or not there is a
pilot. Okay? You can build a B-21 bomber that is flown
autonomously, and I am fine with that, okay? So I am--sorry,
there was a misunderstanding there. That is not what I am
saying at all.
Mr. Gimenez. I am sorry, I am sorry.
Mr. Smith. Yes.
Mr. Gimenez. I am sorry I misunderstood you.
Mr. Smith. That is --
Mr. Gimenez. I just think we could sell that pretty easily,
okay, to the American people.
So it is critical that we win the race and we always have
the most intelligent artificial intelligence if that is the way
we are going to go, and I think that is the way that we are
going to go.
Actually, Ukraine is our Spain, right? The 1936 Spanish
Civil War, the--it was a proxy war, Germans against the
Russians and--you know, and then we--they figured out what the
next war was going to be like, and then they--you know, they
kind of adjusted accordingly. Maybe we didn't in the United
States, it took us--it took Pearl Harbor for us to wake up to
the fact that we were about to go into war, and that is why we
were so far behind. I don't want to be caught like that.
Somebody talked here about the need for drones, and that we
could only produce X number of drones per year. What is China's
capacity to produce these drones?
Let's say Mr. Valentine.
Mr. Valentine. Yes, Congressman Gimenez. First off, I
trained the human pilot that competed against his AI. So it is
probably all my fault that the AI won.
[Laughter.]
Mr. Valentine. But anyway, I don't know the actual capacity
that China can put out right now, but I know that they are
pumping out at least 10,000 a month because that is what
Ukraine is replenishing their stocks with, and that is just one
buyer. They are worldwide, they are subsidized by their
government.
We at Skydio are ready to compete toe to toe with any
company in the world, but it is really hard to fight an unfair
fight when we are fighting an entire country.
Mr. Gimenez. Right.
Mr. Valentine. But I don't know their actual capacity.
Mr. Gimenez. In the Black Sea around Ukraine and parts of
Russia and all that, is the Ukrainian--are the Ukrainians--I am
not going to say Ukrainian navy. I am going to say are
Ukrainians having success in denying the Russians access to
certain parts of the Black Sea?
Mr. Tseng. I think the Ukrainians are having a number of
tactical successes which are leading to strategic dilemmas for
the Russian Navy. And their use of drones and low-cost, one-way
attack drones or one-way missiles, whatever, cheap missiles,
whatever you want to call them, have put the Russian Navy to a
massive--it is a strategic dilemma for them. It is why they
have sunk--why the Ukrainians have sunk so many of those ships.
Mr. Gimenez. So --
Mr. Tseng. They say big ships are big targets in Ukraine.
Mr. Gimenez. So the Ukrainians must have gobs of submarines
and destroyers and frigates and all kinds of stuff, huh?
Mr. Tseng. No.
Mr. Gimenez. They don't?
Mr. Tseng. No.
Mr. Gimenez. How in the world are they doing this? How in
the world are they actually, you know, denying the Russian
Navy, which has gobs of submarines --
Mr. Tseng. Right.
Mr. Gimenez. --and frigates and all kinds of stuff, and
destroyers, and cruisers, how are they doing this?
Mr. Tseng. They are employing asymmetric capabilities that
have tremendous tactical advantage.
And I want to be clear. I think you need a high-low mix.
Again, I am not for a 100 percent--you know, at least in the
next couple of years--drone military.
Mr. Gimenez. Look --
Mr. Tseng. But I want--yes.
Mr. Gimenez. Look, I am not saying you are.
Mr. Tseng. Yes.
Mr. Gimenez. I am just making a point.
Mr. Tseng. Yes.
Mr. Gimenez. Okay? Because I am sure that, if we asked our
military, hey, we want to deny the Russian Navy access or free
reign of the Black Sea, they are going to tell us how many
nuclear submarines they are going to need in order to go in
there to do that, because that is the way they think, all
right?
And if they want to do X, Y, Z, how--we have to ship the--a
carrier task force, you know, over there, which I think--I have
said it many times--we have way too many eggs in too few
baskets because they don't think differently. I would rather
see us disperse our assets so that when one ship is sunk, the
entire battle group is actually useless, and 85 airplanes go
down with it.
So, you know, I am intrigued by the testimony. I believe
that the problem lies--there is a two-pronged problem. It is in
DoD, it is in the Pentagon, and it is in us, right here.
Mr. Tseng. Right.
Mr. Gimenez. Because, because of us, we foster that
culture, that risk-averse culture, the one that wants to go to
what is tried and proven, and I will be promoted, you know, if
I don't rock the boat. Right?
And so I guess I am over. Thank you, Mr. Chairman, for not
shutting me down, and I guess I will have--I will have a second
round. Thank you.
The Chairman. I thank the gentleman from Florida. The chair
now recognizes another local Californian, Ms. Jacobs.
Ms. Jacobs. Yes, though I am from the other side of the
state.
Although Mr. Tseng failed to mention the most important
part of his company when he was doing his intro, which is it is
headquartered in the best congressional district.
[Laughter.]
Ms. Jacobs. So, you know, first, thank you all. I wanted to
ask you, you have all sort of talked in various ways about how
to get--change the way we do contracting and procurement,
getting more towards this problem-based approach. There have
been a few of these different initiatives that have already
been coming out of DoD, and I just wanted to get your all's
take on if you feel like they are going in the right direction,
if you feel like they have helped.
So first I want to ask about Replicator. Do you believe
Replicator has provided the sufficient clarity on its
objectives and plans for companies in--that--particularly ones
who are potentially less familiar with the way DoD generally
works?
And just how do you feel the Replicator program has been
going so far, for any of you who want to answer?
Mr. Jenkins. I can speak briefly to that. So Saildrone is
not a part of Replicator. They have a very, very narrow focus
on kinetic explosive devices for a particular fight near
Taiwan. So it is a very--I am not saying it is the wrong
mission, I am just saying it is a very unique and very narrow.
So it doesn't go much further than that. I don't think anyone
on this panel is actually involved with Replicator, I am
guessing, significantly because of that very narrow focus.
My point is that there is many more things we--challenges
we face around the world that have a very different set of
requirements to the Taiwan Strait challenge, so --
Ms. Jacobs. But would you use the--the way you do
Replicator as a model to say, like, okay, this is another
challenge, but would you advocate doing similar to Replicator
for that challenge, as well?
Mr. Jenkins. It is great to see pace. Speed is very good.
But unless there is recurring program money to take that on and
back it up, it is going to be wasted because it is a small
blip, not a long, continuing story.
The other thing we are seeing, my counterparts in Ukraine
that I am talking to, they are telling me there is a six-week
cadence of having a new solution to that solution being
redundant because they have got a countermeasure. So we have to
be very, very careful when we buy things en masse, that they
are actually going to even work when they are fielded, if they
are fielded. So it could be a very, very large number spent on
something which never actually gets deployed.
Ms. Jacobs. Got it.
Mr. Valentine. Yes, Congresswoman, I think of Replicators--
two things. And on one side it is a strategic concept. And as a
strategic concept, I think it is actually a seed of getting to
this idea of problem-based acquisition that I think we are all
seeking. And of that, this idea of a future battlefield that
has thousands, hundreds of thousands of autonomous systems
supporting a smaller number of humans, I think, is absolutely
the right way to go.
The second part of Replicator is that of an acquisition
program. And there I don't think that the action has met the
rhetoric. So I think the seed is there to create this idea of a
problem-based acquisition system, but what we have actually
done to date on the acquisition side, I don't think has lived
up to that.
Ms. Jacobs. Got it. And then, you know, over the last five
years we have seen a number of different defense innovation
organizations across OSD [Office of the Secretary of Defense]
and the different services. How have you guys felt those have
worked?
Is it too many? Is it not enough? Have you gotten the--what
you need? All of that.
Mr. Tseng. Ma'am, I think all of those--all those programs,
all those institutions, they are Band-Aids. They are Band-Aids
that don't address the root cause of the problem, which is--the
root cause of the problem is all the money in the DoD goes to
the--things that are requirements. And so the Band-Aids are
nice, right, if you are bleeding out and, you know, you want a
Band-Aid, there pretty immediately. But if you have a cancer,
you need to tackle the internal cancer if you want real
fundamental, institutional change.
Ms. Jacobs. Got it. Thank you. And then I wanted to ask
you. Like, to me, one of the things that we are really
struggling with is that, if you look at a lot of the innovation
in the 1950s, 1960s, 1970s, it was a lot of research and
technology that government or DoD initially funded and then
spun out commercially. Now we are in a world where most of the
innovation and research is actually commercially funded, and
then we are trying to figure out how to spin it back into DoD,
which is a slightly different problem set than what--than the
way the system is currently set up.
And I guess one of the ways--if you guys have thoughts on,
like, part of what is hard about that is that, for instance,
the accuracy level you need on AI for something commercial is
very different than the accuracy level you need if you are
using an AI technology to do a kinetic strike, right? Because
then we need, like, 100 percent accuracy. We don't really have
a lot of room for error in a way that you might in a commercial
system.
So how are--how should we be thinking about the way, then,
to take commercially-produced innovations and figure out how to
bring them into DoD, which is a completely different problem
set than the way most of these acquisition processes have been
designed?
Mr. Sankar. I think the software acquisition pathway is a
very powerful tool. I think it has not been used much so far,
and it should be used much more.
With regards to the relevance of commercial technologies
for DoD, right, in our own business we do a lot of work with
autonomous cars. And so you actually do deal with many of the
same types of challenges that you would deal with in a defense
context. Like, it is unacceptable for a car to hit a
pedestrian, but those same types of technologies are also used
in many types of weapon systems.
But we highly encourage the use of the software acquisition
pathway, and using that mechanism actually with groups like the
DIU has been quite beneficial to us.
Ms. Jacobs. The last question I have is somewhat around
some of the language that is in this year's NDAA but, you know,
more broadly about how do we think about the sort of balance
between making sure you guys have your proprietary stuff, but
then making sure DoD can, for instance, repair. We have got
some right to repair language in this year's NDAA--can actually
provide oversight.
So, like, how should we be thinking about that question of,
like, how much we can let you guys have a black box versus the
need for us to be able to repair or do oversight or make sure
that things from different companies are actually able to
operate together?
Mr. Sankar. To address this from the software perspective,
people tend to think about vendor lock. You know, any decision
my engineers make, they are locking me into that technology.
The real question is not do I face vendor lock in my own
software stack, it is what is the switching cost? If the
switching cost is two days, I really don't care. They can go
forth and make whatever decisions they would like. If it is two
decades, I care a lot. I am going to have a huge amount of
scrutiny over the technical decisions they are making and the
implications of that.
And I think if we could just own it--so in particular, a
lot of the time the DoD says, well, I want to avoid vendor
lock, so I am going to build it myself. They are just locking
themselves into this, you know, Galapagos Island that doesn't
benefit from any commercial innovation or R&D. They are not
going to be able to outspend the commercial world. But instead,
if we started thinking about strategically where do we have
vendor lock that has unacceptable switching costs, we can bring
market forces to that.
Why does the commercial world not face this? Because the
incentive for every company is to compete based on switching
costs. So people invest in making that possible, and I think a
similar approach is very, very possible within DoD.
Ms. Jacobs. Thank you, Mr. Chairman, for the extra two
minutes. I am getting spoiled over here.
I yield back.
The Chairman. The gentlelady yields back. The chair now
recognizes Ms. Mace of South Carolina.
Ms. Mace. Thank you, Mr. Chairman, and thank you all for
being here this morning and briefing us.
Mr. Tseng, you mentioned GPS a couple of times in your
testimony and answers. How do we address the GPS issue? How
would you solve that problem?
Mr. Tseng. I think about, one, putting ourselves in real,
relevant battlefield conditions in training, right? In the SEAL
teams and the entire military we say, you know, the more you
sweat in training, the less you bleed in war, and that is 100
percent true.
It is very disappointing. I think I can count probably on
one hand, maybe on two, but I am, like, pretty sure on one hand
the amount of actual relevant electronic warfare pre-deployment
workups that our troops go through. They are simply not
experiencing the problem, they do not know what will work or
what doesn't work when it comes time to deploy, and that is
something that I think needs to be changed immediately.
Ms. Mace. But also in terms of GPS, I mean, China is trying
to hack our GPS systems. Like, what is the future? What do we
utilize? How do we get away from GPS, I guess, is where I was
trying to go with that.
Mr. Tseng. Yes. I think you have to build systems that are
completely non-reliant on it. It is not going to be a reliable
support mechanism on the battlefield.
And in humans, right, we--in the SEAL teams, fighter
pilots, you learn to navigate without GPS, right? But
obviously, for the weapon systems that have been proliferated,
the surface-to-air missiles --
Ms. Mace. If you are flying a C-17, right?
Mr. Tseng. Right, it is--you don't want to go up against a
surface-to-air missile system, right?
Ms. Mace. Right.
Mr. Tseng. And so you need to build systems that
fundamentally don't rely on GPS.
Ms. Mace. Mr. Valentine, did you want to jump in?
Mr. Valentine. Yes, Ms. Mace, the way we have attacked
that, we fundamentally believe that AI, which many of the
companies at this table represent, is the antidote to GPS
jamming and the GPS spoofing.
Ms. Mace. I agree.
Mr. Valentine. So the way we have done that is we have
taken navigation cameras that were designed to prevent the
drone from running into things, and realized that we can use
those to do a visual-based navigation system, much like an old
Tomahawk would navigate by comparing two pictures, realizing
that, oh, I should be here, but I am here, and then doing some
trigonometry. We are kind of doing the same thing, but we are
doing it roughly 60 times a second through about 8 neural
networks. And so it is much more accurate. And we have seen
great success with that, to the point where after a 40-minute
flight of jamming against GPS and active spoofing with no GPS--
our drone, by the way, in the rain in Ukraine--came back and
landed about 20 feet from where the operator was. So that was
pretty impressive.
So we fundamentally think that the technologies that we
represent at this table will ultimately solve that GPS
challenge.
Ms. Mace. Mr. Ludwig?
Mr. Ludwig. I would really like to emphasize that this is a
software problem, not a hardware problem. And we have software-
defined radios today, and we also have very sophisticated test
infrastructure for being able to test these things.
I think especially when we talk about a system like GPS, we
think of it as satellites and electromagnetic waves.
Fundamentally, though, the way that these autonomy systems work
in the presence of electromagnetic interference is very
sophisticated software, which gets back to this splitting the
software and the hardware components and the importance I think
that has for the DoD.
Ms. Mace. Okay. And then you mentioned too, Mr. Ludwig,
since you got the microphone, in your testimony about China
leapfrogging the U.S. in terms of autonomy. When will they
leapfrog us on autonomy?
Mr. Ludwig. I think this has already happened. A group of
our leadership team earlier this year was at the Beijing
International Auto Show, and we got to test drive the latest
commercially-available Chinese vehicles, literally going to
dealerships in China and trying those systems. And I think it
is fair to say that the Tesla autopilot system is actually not
competitive compared to the things that we saw in China. Some
of the most impressive technologies coming directly from
Huawei, in fact.
Ms. Mace. Okay. And then, in terms of technology and AI,
when does China leapfrog the U.S.?
Mr. Ludwig. I think that is harder to say, but I think it
is important to note that the innovation is very vibrant in
China. The most recent CVPR [computer vision and pattern
recognition] conference, which is the world-leading Computer
Vision conference, had more submissions from Chinese
researchers than any other nationality.
And so the amount of innovation that is happening in China
is really quite incredible, and we should take it very
seriously.
Ms. Mace. Okay. And then speaking of innovation--and I
agree with everything, just about everything that has been said
today--I mean, we have created a bureaucracy, it takes a really
long time. Things are changing so rapidly, and you guys have
testified today on how quickly the technology and the
capabilities of your technology are changing, as well.
Short of World War III, which nobody in here wants, we
always want to avoid that, what do we need to do? How do we
force innovation? How do we force ourselves to be more nimble,
understanding that we are not going to get this massive
overhaul?
I agree with you on the problem-solving acquisition sort of
process. That is very private sector. That is not the
environment that we have right now. So given the chessboard
that we have our pawns on, that the chips are on, how do we--
what small parts will make a big difference?
What are some small things we can do now to be more
innovative, to be faster, to be more nimble?
As you all say, the pieces are in place, but we are just
not doing it. It could--personnel could be the answer. It could
be technology. But how do we incentivize?
I mean, I have seen some really crazy things. I had a
hearing, I think it was earlier this year or last year. It was
like $300 million spent on a software system by DoD. Never got
implemented because there are--none of the service branches
were ready for implementation. We basically poured gasoline on
$300 million and lit a match. Just never--you know, so how do
we avoid that?
I mean, I just--what is the answer here? What can we do now
in the next 12 months with the regulatory environment, the
acquisition environment we have? What can we do now to make a
bigger difference?
Mr. Sankar?
Mr. Sankar. I go back to the submarine-launched ballistic
missile. When Admiral Rayborn was developing that in the 1960s,
we had four competing programs going on simultaneously.
I think the root cause for the lack of innovation is that
when you have a single source that gets to decide whether they
want the innovation or not, they don't. There has to be
something worse than change, which is irrelevance, in order to
get people to adopt and change faster. And so I think we need
more competing initiatives.
You know, we struggle for a lack of competition inside of
government. We are often trying to externalize that competition
to the companies, but that is not the root cause of the issue.
It is--you know, this is why I think empowering the COCOMs does
matter, because it provides incremental signal on different
needs and incentives of where to go that forces services to
react to those needs --
Ms. Mace. But how do we break up those monopolies? Like,
how do we break it up? What can we do now, in terms of the
process, the environment we have today?
Mr. Sankar. Right, I think a very tactical solution is
every program that is a major program should have multiple
competing PMs.
When we were going through the Titan program with the Army,
there was one PM who was overseeing the competition between
Raytheon and Palantir. Maybe we should have two PMs, one
attached to Palantir, one attached to Raytheon. And those PMs
wake up every day trying to win, trying to beat each other. It
provides incentive on the margin to adopt new technologies, to
go faster, to trade off small requirements, you know, use human
judgment to see what is going to happen, and that then puts
natural weight into the COCOM command.
You know, what does General Flynn in the Pacific think he
wants? Does he want this Titan or that Titan? And how do we get
into constant iterative cycles? I think it provides the right
basis. It starts to approximate what commercial competition
looks like.
Ms. Mace. And then, if the others--if everyone else--I will
start over here, Mr. Valentine, and we will go through the
table.
Mr. Valentine. I agree with Mr. Sankar. Competition is an
incredibly important part of this. I also think personnel, that
is a big part of it.
There are several programs the Department already runs,
where they do fellowships with private-sector companies so that
program managers, so that military officers can start to
understand what it means to actually innovate. I think we
should fund those and send more people to them.
Ms. Mace. Mr. Tseng?
Mr. Tseng. I think you can create emergencies around the
problem. I think everybody in the DoD, I think leaders in
Congress recognize the state of our national security, the
state of our technological edge is in an emergency situation. I
say ``emergency,'' because I think Congress and the DoD does a
very good job when an emergency has been declared, right? And
that is when, for whatever reason, we all get the most bang for
our buck as taxpayers.
But that is what I think about. You can create these mini-
competitive Manhattan Projects, where you are funding a handful
of these efforts to create these capabilities and technologies,
but there has to be a sense of urgency behind it. There has to
be a recognized--a recognition that we are in a very
challenging situation.
Ms. Mace. Mr. Ludwig?
Mr. Ludwig. And I would just underline that I really do
feel the primary innovation, looking forward to the next
century, is going to be in software innovation. And so right
now, as a matter of fact, the DoD does not purchase much
software. And I think that, in order to unlock that huge amount
of innovation in software, there just have to be more
mechanisms to more iteratively and more quickly procure
software.
Ms. Mace. Mr. Jenkins?
Mr. Jenkins. For me it is all about incentivization. How do
you incentivize companies to innovate, innovate competitively
amongst each other? Then how do you incentivize budget creators
to be bold in what they are choosing? If you can do those two
things, there is a huge amount of private capital, as this
panel demonstrates, that can put their power to bear on the
solutions. But if there is no pathway, no roadmap to get to the
end goal, it is not going to happen. So incentivize private
companies, incentivize government to choose the right products.
Ms. Mace. Thank you, Mr. Chairman, I yield back.
The Chairman. The Chair now recognizes the gentleman from
Texas, Mr. Veasey.
Mr. Veasey. Thank you, Mr. Chairman, and I thank the
panelists for being here today.
Mr. Tseng, I wanted to specifically ask you about the V-
BAT. And I know that you said that they may not directly fit in
with the Air Force's Collaborative Combat Aircraft Program,
which focuses on AI-piloted jets working alongside crews like
in the F-35, for instance. And so I wanted you to elaborate
that--elaborate on that a bit, and just explain how these V-BAT
teams could potentially be integrated with such a program in
the future, and what are some key considerations for making
such integration successful.
Mr. Tseng. Sure, absolutely. The V-BAT, at its core, the
problem it is solving is that of a targeting problem. A long--
you need long endurance. You need long duration to actually
target assets on the battlefield. We learned this very well
during the global war on terror, and we utilized a wide array
of drones to execute these targeting cycles. And so the V-BAT
is a platform that is really a next-generation targeting
platform that is meant to solve that targeting problem.
The Air Force CCA [collaborative combat aircraft] program
is a--it is an air dominance, air superiority program of record
that looks to solve the problem how can we establish air
superiority? And so these two things are in concert when I
think about them. Where you would say, you know, the CCAs that
are being contemplated are primarily strike platforms, well,
what the V-BAT does is it provides targets for those platforms
to strike.
Mr. Veasey. Okay. So given that the Air Force's focus on is
on developing these AI-piloted jets, how could a V-BAT--because
I know the F-35 has come up--how can V-BAT teams be
complementary to F-35s?
And how can Shield AI's technology enhance the
effectiveness of the F-35 in joint operations?
Mr. Tseng. Yes, a handful of ways.
The V-BATs themselves, they can forward deploy, they are
highly mobile. You don't need runways to launch out of them.
They are logistically simple. And so you want to start creating
that targeting picture well ahead of the utilization of any
long-range--like strike assets. Whether they are missiles,
whether they are fighter jets, you need the targeting picture
first.
In the SEAL teams, in Special Operations Command, it was
driven into our head that 99 percent of, you know, what we do
was intelligence operations, 1 percent is the kinetic piece.
Obviously, a very important piece, but you have to know what
you are going to hit, when you are going to hit it, how you are
going to hit it, and that is the targeting picture that the
U.S. has become, you know, best in the world, premiere at doing
it.
V-BAT is well out ahead. I am talking days, weeks, months
ahead. Creating that targeting picture is how I contemplate--in
these electronically warfare-contested environments, with--
while GPS is being jammed, while communications is being
jammed, is the role that it plays directly into our air
superiority, you know, mission with a joint strike fighter,
with a CCA, et cetera.
Mr. Veasey. Yes, no, thank you very much.
And Mr. Sankar, you said something that stuck out a little
bit for me. I know that in order to make all of these changes
and to be able to improve these procurement processes that you
talked a lot about in your testimony, to me it sounds like that
is obviously going to take, you know, resources to be able to
do that. And I know that also there seems to be a bullishness
not just on the panel, but for the American public to see more
of these, you know, drone-type capabilities, these things that
don't require as much human risk involved. And, of course, we
hear that a lot from our constituents.
But being able to do that, it seems like it would take,
like, a generational investment to be able to do it. It just
doesn't happen overnight. If you think about, you know, what
this country did in the 1980s, we spent a lot of money in order
to create this perception that the U.S. was the biggest
military power in the world, and we were. Not only did we spend
money domestically at our defense plants, but we put base--
forward bases around the country.
Do you think that, A, that we are able to do all of these
things that you would like to do without spending much more
than we already do?
And also, do you think that just having these more, you
know, AI-type capabilities or, you know, planes that don't have
pilots, that that is going to really be enough in order to
deter war? Because on this panel, that should be the first
thing that we, you know, work on.
I know that, obviously, we want to make sure that we are
being competitive with the Chinese, but we obviously want to
have the perception and have enough military might and muscle
and have other people see that we do in order to deter war. And
so does that do that?
Mr. Sankar. I absolutely--well, you could say that we are
spending at historic lows as a percentage of GDP [gross
domestic product] relative to what we have in the past.
But I do actually think that the primary issue is of asset
allocation. If we said we have the budget we have, are we
allocating it to maximize our lethality, I think the answer is
no there. And I think a big part of that is how much innovation
are we able to benefit from?
This decoupling of the commercial world from the defense
world is very profound. When at the beginning of World War II,
when the U.S. Army went to Pontiac and asked them to start
making anti-aircraft guns, the U.S. Army's production time per
unit was 3.5 hours. For Pontiac, it was 15 minutes. Why was it
15 minutes? Is it because they are smarter? No, it is because
they had so much experience making cars that they were able to
transfer laterally here.
Only 30 to 40 percent of Chinese primes' revenue comes from
the PLA [People's Liberation Army]. So that cheap toaster your
neighbor is buying on Amazon is subsidizing lethality against
U.S. service members. This is why it is such a profound problem
that 86 percent of defense spending goes to defense-specific
companies that have no ability to benefit or amortize or learn
any lessons from the broader commercial market.
You know, this is a challenge for us here. Actually, sorry,
it is 94 percent. And it didn't used to be that way. This is a
consequence of winning the Cold War. When we were competing,
trying to win the Cold War, we had this diversified industrial
base. Why is that important? Today it is very hard for me to
compete on price. I would love to go to existing programs and
say, I can do this at half the price. Absolutely no one is
interested in that proposition, which is crazy, because that is
exactly how I built my commercial business. I go to companies
and compete on price. So how do we create that mechanism?
The other--one of the reasons that is incredibly difficult
is we have this fixation on the cost-type contracting. I want
to pay you for what it costs. Well, I want to spend billions of
dollars to create technologies that mean I can deliver for you
at half the price. I can do that because I am going to have
twice the margin. And so we need to move to a value
measurement. What is the government getting for this, as
opposed to what did it cost me to build the thing? And by doing
that, you are going to liberate massive amounts of the budget
to reinvest in the capabilities that drive incremental
lethality. Unfortunately, that is what our competitor is doing.
We have no time to waste on that.
The Chairman. The chair now recognizes the gentlelady from
Virginia.
Mrs. Kiggans. Thank you, Mr. --
The Chairman. Mrs. Kiggans.
Mrs. Kiggans. Thank you, Mr. Chair. We have talked a lot
about drones and innovation in modern warfare. Today, and much
like machine guns in World War I, small drones have emerged as
the new staple of the modern battlefield. This is demonstrated
by the ongoing conflict in Ukraine, where drones are used in a
wide range of applicability, ranging from FPV drones being
utilized to kamikaze strikes or larger quadcopter drones being
used for ISR applications. Interestingly, NATO [North Atlantic
Treaty Organization] recently reported that two-thirds of all
Russian vehicle losses in the conflict have been a result of
Ukraine's employment of FPV kamikaze drones.
I know that you all are probably watching just some of the
challenges that Congress has had with Ukraine aid and funding,
and also just listening to you all talk about how the DoD
budget and acquisition process has just really stifled your
ability to develop new technologies. So could you address for
me just how impactful those--that Ukraine aid piece was for
your companies, for your--improving your business models,
expanding your portfolios, and just developing new technology?
Mr. Sankar. For me, the Ukraine--for us, we have gotten
zero resource from the Ukraine aid that has been given. We have
been on the ground there since May of 2022. We have a full
office in Kyiv. We were out forward. We are there at our own
expense to learn. And the rate of learning there far exceeds
what we are often able to learn in any other theater that we
are at.
Now, you know, the Europeans have contributed resources,
but I think it is kind of a shocking statement that we have not
had any contributions from the resources that have been given
to the Ukrainians.
Mr. Valentine. Congresswoman, also of the roughly 1,000
systems we have in Ukraine right now, I think the vast majority
of them were either donated by us or were purchased by the
donor community at the beginning of the conflict. So I can't
directly trace any of the Ukraine funding to supplying
systems--from our company, at least--to Ukraine. But I can
absolutely tell you the need is there. They have the need, they
have the will, and they have the capability to use this, as you
mentioned, to a tactical and strategic advantage. At this point
they just need the stuff delivered.
Mrs. Kiggans. Thank you. And clearly smaller drones are
here to stay. And the DoD needs to examine ways it can
strengthen the supply chain of domestically-produced drones. So
Mr. Valentine, can you tell me what ways the DoD can help to
foster a strong domestic supply chain of smaller drones?
Mr. Valentine. Yes, Congresswoman. Again, this comes back
to when we start to buy things consistently. And I think Mr.
Tseng actually said it correctly, and Ranking Member Smith, as
well.
This isn't give us a contract for 10 years. But if we can
demonstrate that we have a consistent set of small bites at the
apple, and we will iterate and deliver whatever capability you
need, but once that is demonstrated, now we have the essential
buying signal we need to go and buy all the constituent parts.
There are things that take a long time to get, and sometimes
those things can take four months, which has led to this idea--
at least that I have started to think--about around stockpiling
so that we have some tremendous demand shock. We can flash the
systems with whatever the latest software build is and get them
out the door to be able to respond to that demand shock.
And then the only other thing that I would mention are
chips are always an issue. The more that we can bring that
chip-making capability, at least to friendly countries, ones
that we have access to during a time of conflict, I think that
is supremely important.
Mrs. Kiggans. Thank you.
And Mr. Tseng and Mr. Sankar, you guys talked about just
the--moving from the requirement space to--well, Mr. Tseng
did--moving from a requirement space to problem-based, you
know, acquisition process. And so I am just curious. How do you
all, as software companies, work with the primes to integrate
the latest and best technology?
And then how do you work with the warfighter to actually
hear from them what they need?
Mr. Tseng. Yes, we work with the primes every single day.
And it comes to, actually, a lot of people's recognition, the
primes' recognition, the warfighters' recognition that the best
way to solve this problem is with the software-first mentality,
things that both my--you know, all of my colleagues here have
referenced.
And the way that we, you know, hear from them, work with
them, no different than any commercial company, you go out, you
find your customers, you talk to them, you aggregate as much
input as possible, you filter out the noise, you see--you
recognize what makes sense, which is standard best product
management practice at any, you know, Fortune 500 company. That
is the model that we take. And I--you know, yes.
Mrs. Kiggans. Go ahead.
Mr. Sankar. And I would just add to that, you know, we ship
out 90,000 software upgrades to our fleet a week.
One of the really interesting lessons for me during the
Afghan NEO [noncombatant evacuation operations] was that
usually in a crisis, when something happens, the old mentality
is that you freeze things. You know, the change seems risky.
But actually in the crisis, we accelerated the amount of change
that was happening. We pushed out more than 90,000 upgrades a
week because it enabled us to be responsive. Software became
the most malleable weapon system.
So I can't understate the importance of being out there
with the war fighters, getting the--each incremental piece of
feedback, and then closing those code chains. We spent a lot of
time talking about kill chains. If you want to close kill
chains, you need to close the code chains as quickly as
possible.
Mrs. Kiggans. Is there a way that Congress can intervene to
improve that process of communication between the warfighter,
or do you feel pretty good about that, then?
Mr. Sankar. I would come back to my suggestion of giving
more authority and voice to the COCOMs, that that is where the
need really is, and that is going to drive the speed of these
code chains.
Mrs. Kiggans. And out of curiosity, what do--do you speak
to the pilots out there? You are talking to--this is the pilot
end of the table. And just about the--this progression of
making us--of us just not relevant anymore, and do they
recognize that? What--how do they feel about that?
Mr. Sankar. They feel good about it. I tell people, look,
being a Navy SEAL is really cool until you are asked to fight
inside a tunnel system, right, where attrition rates, casualty
rates are 80-plus percent.
Being a pilot is really, really cool until you are going up
against an integrated surface-to-air missile system, where you
have a 0.1 percent chance of survival. That is not a mission I
want any of our pilots ever to take.
Mrs. Kiggans. I agree with you. As the mother of a future
naval aviator, I agree 100 percent.
Thank you very much, and I yield back.
The Chairman. The chair now recognizes the gentleman from
Georgia, Mr. McCormick.
Dr. McCormick. Thank you, Mr. Chair, and I am sure AI will
never be able to duplicate our good-looking pilots on the
beach, which is really important for movies.
[Laughter.]
Dr. McCormick. I am really enjoying this. This is the first
field hearing I have ever participated in. This is fantastic.
It is a fantastic panel, a very fascinating topic that really
is maybe one of the more important, germane topics for what we
need to conquer in the near future, because it will affect our
readiness for our men and women and also for our great country
because, let's face it, we don't have enough qualified people
out there to even man the 17 naval vessels that can't be
deployed this year because we have an under-manning process.
And the more automated we become, the fewer people get killed,
the better we do our process, the less we spend on recruiting
instead of R&D. I think this is fantastic topic. So thank you,
Mr. Chair, for doing this for us, and thank you to the panel
for taking time from your very important responsibilities for
doing this, because I love this.
I love the idea of the First Breakfast. I thought that was
a--just spot on. I might actually watch this hearing over
again, just because I am learning so much after all of you
speaking.
In medicine, we have also had that same resistance to
change. Because I am not just a pilot, I am an ER [emergency
room] doc, in case--you ever want to know if there is a Marine
in the room or a pilot in the room, how do you know? They will
tell you, right?
[Laughter.]
Dr. McCormick. But in medicine we resist change, too. At
one time it was considered a great thing to have dirt all over
your jacket. It shows how salty you were. And they actually
made fun of people like Dr. Lister, who wanted to use
antiseptic, or somebody who actually wanted to wash their hands
in between surgeries. We resist change, even when it is a very
good thing for the future.
As a matter of fact, we demonize people, as we have pointed
out, for people who have done the right thing in the military,
and in medicine, and everything else like that. We always
resist change. But thank God for change, because that is what
is going to keep us relevant into the next generation of
weapons systems.
I was kind of trying to figure out who is more answerable
to what we are actually talking about today. Are we answering
to you, or are you answering to us as far as how do we change
this? Because ultimately, we are the people who are going to be
the instrument to changing how this process works so that you
can do your job better. But ultimately, we are answerable to
the people of the United States because we are taking their
money and funding it, which brings up an interesting point.
This is the first time, maybe since Roman history, where
you have individuals that are as powerful, maybe even more
powerful, than the government. You remember Pompeii. He could
raise his own army, he could fund his own consulship. Those
days have been gone for a long time ago. It used to be that the
United States and Russia were the only two countries that could
put people up in outer space. Now you have an individual who
can put more spaceships up in outer space than the entire
American Government.
Times have changed. Seventy-five percent of our government
is based on welfare programs. Another 10 percent, we argue all
of our non-discretionary spending--it is what we spend
everything else on, and we think that those are the powerful
committees, whereas I would say that the last bastion of hope
for America and its progress in technology, really, is right
here in this room.
The Armed Services Committee is the only real investor in
progress in technology, if you think about it. Eight hundred
billion is being spent, not only in the advancements as we are
talking, we waste a lot of money. But if you think about it,
what else do we invest in anymore? We don't. The American
Government is not good at investing. We need this hand-in-hand
attitude, where we actually have corporations doing the hard
work for us so you can make a profit, but we can benefit.
Because let's face it, NASA [National Aeronautics and Space
Administration] is not the one going up to outer space to
rescue people from a space station that is 30, 40 years old.
That is where we are at. That is the reality of where we are at
right now. So I am really excited about what you guys bring,
especially when we look to the future of budgeting.
You talked about stockpiling. It worries me a little bit,
because I agree with the chair. You become antiquated very
quickly. We have watched this in our weapons systems even now.
And you talk about--not to pick on F-16 pilots or F-16s in
general, but it struck me as curious that Ukraine even wants
them when you consider about the future of warfare, and what
you can do against an F-16--and you only have a few, compared
to the SAMs [surface-to-air missiles] and everything else that
you have in Russia. Why are we even bothering with that
technology?
Let's stick to the more advanced technologies, which also
worries me about the future of countermeasures, where you have
Iran and China supplying our enemies with very cheap,
technologically-advanced weaponry. And yes, we can--the kill
chain book alluded to this, and we have been briefed on it,
right, how we are countering those measures.
But how--here is my first question to you guys is, how do
we come up with new technologies to counter these drones that
are going to come after us and our legacy systems?
And I think we do need carriers, but I am worried about
what do we need to be investing in to counter the measures that
we are actually developing from Iran and China and things that
are going to protect our larger systems?
And I will start with the First--I just love the First
Breakfast comment, so we will start with you, Mr. Sankar.
Mr. Sankar. Well, thank you.
I will say, when we talk about stockpiling, we tend to
focus on the stock in the pile. But what really matters is your
ability to produce the stock. Innovation is a consequence of
productivity. When you don't produce, you can't innovate on
what you are--what it is that you are producing. So we have to
be thinking about what are we buying here?
We are not actually buying the end product that you are
then going to stockpile because, you are right, it is--there is
an ephemeris component to that. You are actually keeping alive
the knowledge and the innovation cycle. You are funding the
OODA loop. And I think that is a critical part of how are we
going to counteract these capabilities. Because, you know, it
is not a--it is a moving target itself. Like, as we counteract,
they are going to counteract again, and we have to stay on top
of that.
So I think the broad set of investments comes down to the
need to produce enough scale of mass so that we can keep
innovating on what it is that we are producing.
Dr. McCormick. And Mr. Tseng, did I hear that you--I mean,
we talked about the legacy systems. We talked about different
things we use that aren't effective in Ukraine. Did I actually
hear you say that the HIMARs were not a good weapon in Ukraine?
Mr. Tseng. GPS guided rounds.
Dr. McCormick. Okay.
Mr. Tseng. There are certain rounds. Again, it is a mix of
some things that work and some things that don't. And if it has
GPS, it has got a very, very, very, very, very low success
rate.
Dr. McCormick. Okay, and I understand--I have heard from
different companies that we are not delivering the weapons that
we actually have stockpiled in America to the Ukraine basin,
and I am not sure if that is due to purchasing or delivery
problems or regulations.
What is the biggest problem in our deliverable? Because we
have the weapons that we are not giving to them that they could
use in the battlefield that are the most effective weapons I
have seen. You could talk about switchblades. You could talk
about any sort of drones. You could talk about all kinds of
different things. But why are we not delivering? Is it the
purchasing problem or is it the delivery?
Mr. Tseng. It is probably a combination of both. There is a
lot of noise coming out of Ukraine. There is certainly a
bureaucracy going through that process right now in terms of
the foreign aid process that is something that our company is
going through right now with the Ukrainians, hand in hand.
And then certainly, also, when you talk about the delivery
mechanism, again, the Ukrainians very--they force companies to
come to the front line now and--right? And work alongside them
as they use their product. And this is a very hard thing. It is
a very expensive thing to do, as a company. But I give them all
the credit in the world because they simply don't want things
that are not going to work on the battlefield.
And so while there is a lot of noise in this space, I think
you see, if it is not on a buy list from the Ukrainians, there
is a good chance they--like, they just haven't tested it, or
they have tested it and it is not being successful, period.
Dr. McCormick. And then final question, Mr. Ludwig, you
mentioned the importance of software. When we talk about
investing in the future, whether it be software or hardware,
the difference between quantum, AI, and autonomous, those three
things all kind of combine into one system sometimes, and if
they do--if we get those all together, man, it is going to
change the world. And of course, that is why it is really
important to develop these hand in hand.
I don't know if you think that we are going to be able to--
is it realistic that in the next couple of years we will be
able to integrate those three things in a realistic way that is
going to affect the--shape the battlefield?
Mr. Ludwig. Well, certainly, AI and autonomy, these things
are deeply intertwined. And those fields are evolving at a
rapid pace. Quantum still exists more on the research
boundaries. The primary promise of quantum is the ability to
more easily crack cryptography standards. And if--certainly, if
a country does have the ability to crack all cryptography
standards, that is a game-changing capability. But that is more
in the research phases today.
I really want to underline, though, just how quickly AI and
autonomy are advancing and, again, how little, frankly, focus
there is on this, generally, at the Department. We are,
generally, talking about building--talking about systems like a
plane or a boat, and not talking about the advanced software
capabilities that are required to actually make that effective
in the modern war.
Dr. McCormick. Thank you.
With that, I yield, sir.
The Chairman. I thank the gentleman.
I wanted to ask--we talked about the tests and seeing how
things work in Ukraine. Have any of you had difficulty in being
able to sell to Ukraine, any restrictions put on your ability
to sell by our government?
Mr. Valentine?
Mr. Valentine. Congressman, we do have issues getting
systems into country. Many of our systems that have thermal
cameras or they are not ITAR [International Traffic in Arms
Regulations], but they are BIS-controlled [Bureau of Industry
and Security], getting those export licenses can be a
challenge, especially when that export license process starts
with getting an end user to certify here is me, here is my
address, and I need X number of these things. That is pretty
challenging to get from a platoon of soldiers in Ukraine, even
working through the Ministry of Defense. So that can be
somewhat challenging.
Mr. Tseng. I will just say there is not a fundamental,
like, limiting--like, they are--the export regulations have
been solid. Yes, you have to go through a license process, but
it hasn't been a limiting factor in terms of getting
capabilities over there.
The Chairman. Good. Mr. Ludwig, you made reference to the
DIU hedge fund needs to be fully funded. How much is it being
funded? I am skinning my ignorance. I don't know.
Mr. Ludwig. I would actually have to ask my team for the
precise number. I am not sure.
The Chairman. Yes, I would like to know what ``fully
funded'' is for that hedge fund.
And you also made reference to the software acquisition
process. Is it different? Is it different from this
requirements-based acquisition process we have been talking
about?
Mr. Ludwig. Mainly, I am referring to the splitting of
procuring software from hardware. And so often times software
is thought of as a, let's say, the thing that you do after you
have thought of the hardware system. And what this does is this
forces many hardware companies to think about software as a
cost center, rather than a profit center.
And in the reality of the talent in software, it is
extremely competitive, and often times it is very difficult to
hire the talent that is required to build these most advanced
systems if you are thinking of this as a cost center.
The Chairman. Mr. Tseng, a couple of things you said, I
think, are big takeaways from today. One is the requirements
based system is a cancer. I like that, it is very easy to
understand. And another was that we should dedicate 20 to 25
percent of the acquisition budget to a problems-based system,
as opposed to a requirements-based system. Give me another
thing that you think this committee should take away from this
committee, other than--this hearing, other than those two
principles.
Mr. Tseng. Those are very, very top of my list, so I am
glad that that would be your number one.
Behind that, I think it is about really, really being
customer-centric, warfighter-centric here. Again, whether it is
going to the COCOMs, as my colleague, Shyam, talked about, or
increasing the amount of reality in training on the
battlefield.
And that, to me, is what is nearest and dearest, because I
have friends, I have teammates who are preparing for
deployment. And you ask them, you know, are you going up
against what the battlefield is going to look like, the
electronic warfare battlefield? And they tell me--they are
like, no, we are still preparing in the same way we prepared
for Iraq, for Afghanistan. And that, to me, is--that is a
heartbreaking thing, because I think we owe it to them to
actually prepare them and train them for what the conflict will
actually be like.
The Chairman. Great. Mr. Sankar, you have had a lot of very
thoughtful suggestions in here. I am really impressed with your
thought process on this.
The Defense Department is made up of a lot of very
patriotic Americans who love this country and want to help us
be successful. It is also the largest organization on the
planet, and culturally very indoctrinated. And it has been my
experience in 22 years on this committee that it does not
change unless it is forced to change. I offer that as the
backdrop to this question: Do you have some statutory
suggestions that you think the Congress should consider
imposing on the Department to bring about the kind of changes
you think are important?
Mr. Sankar. All my suggestions come around to the idea of
increasing the amount of competition within the Department
across programs. So ideas like competing program managers,
competing PMs to go after this, to create the right incentives
to adopt innovative approaches that shave time off schedule,
that provide lower-cost options. I think we should--we do need
a little more crazy.
I think Congress--you know, in the Rickover story, Congress
doesn't probably get as much credit as it needs for how much
protection it--you know, Zumwalt said, ``The Navy has three
enemies: the Soviet Union, the Air Force, and Hyman Rickover.''
And you can think about how important the role of Congress
played was there.
I think we should recognize that we have uniquely talented
people in the service, and there is probably a role for some
DOPMA [Defense Officer Personnel Management Act] reform here.
You know, the idea that people have to keep moving every three
to four years, even though they might be the most uniquely
suited person to carry some of these roles forward, is a
challenge. It is not how we would operate in the private
sector, right? You know, when you get the--some projects can
only be done by a certain set of humans, and you have to, like,
wrap yourself around them and make everything happen as a
consequence of that. So I think that is pretty important.
And then we should probably be more open--Andy Grove had a
saying. First you got to let chaos reign, then you rein in
chaos. We have very low tolerance in the Department for letting
chaos reign. You know, it is okay for it to be confusing, what
is happening with Replicator. My question would actually be
maybe it should be more confusing. Maybe we need two or three
Replicators happening at the same time to create the necessary
sort of incentives here.
You know, I think space is one of the most innovative areas
because, frankly, it is a bit of a food fight between NRO
[National Reconnaissance Office], NGA [National Geospatial-
Intelligence Agency], Space Force, Spacecom, and that is
increasing our lethality. So you could look at that and say,
oh, this looks like a bug. I see it as a huge feature. You
know, we are going to be better off because of it.
Now, what are the other areas where we don't have enough of
that? Is Ukraine's idea on unmanned surface crazy? It is not
even about whether the unmanned surface is going to work. Does
it provide the necessary incentive for everyone else to adapt
around that? Does it provide the reason, as you said, for them
to change?
The Chairman. Well, I would make reference to what the
ranking member said earlier, and that is this committee, in the
last 10 or 12 years, has given the Department all sort of
authorities. They won't use them.
So I would urge you all and your counterparts, who are not
necessarily witnesses here today but are paying attention to
this, come to us with specific statutory language that you
think will be--have a forcing function. We are all ears. We
want to get it from you, but you are the folks that interact
with them. You are the best person to give us that language.
And I would end my comments with what I think is the most
important thing that you have said here, Mr. Sankar, and that
is we are spending on defense at historic lows, as a percentage
of our GDP. That is a fact for over a century. The last three
years has been the lowest spending on defense as a percentage
of GDP in our country. That is unacceptable. The last 2 years,
we are at 2.9 percent, this year is at 3 percent. That is
inviting problems, and we have got to get after that. We have
got to spend it better. Lord knows, we need to be wiser about
it, but we also have to recognize that we have gone from a
strategy that focused on fighting two wars simultaneously and
successfully and defending the homeland to one war successfully
and defending the homeland. The world is too volatile for that.
And with that I yield to the ranking member.
Mr. Smith. I thank you. Just a couple of closing
observations.
First of all, I love what Mr. Sankar said there. I think
unity is the most highly overrated thing that I have ever come
across. You know, I was, oh, no, we can't have disagreement, we
can't--there are political sides to that that create problems.
No, I mean, you need to challenge assumptions. And a lot of
times, when people are pushing for unity, they are basically
pushing for you to shut up and do what they want, as opposed to
challenging it. So we need to move past that.
I think the readiness point you made at the end is
something I hadn't thought of, that I think our readiness
subcommittee should really take a serious look at: how are we
training our service members right now to go into battle,
acquisition and procurement notwithstanding? I hadn't thought
of that point.
The big thing on the--and I agree with the chairman that we
have given DoD the authority they need to do, to your point, a
problem-solving-based acquisition process. But what we haven't
done is we haven't given them the money. At the end of the day,
that is the real problem. The Replicator, I don't know, I think
it is classified, what it spends, and I don't know. So whatever
number. But it is not in the billions, let's put it that way.
It is a very small amount of money, as are most of the OTA
[Other Transactions Authority] authorities. So we lock people
in in the money. So I really want to sort of get after how can
we free up more money, whether it is 25 percent of the budget
or whatever percent of the budget to get outside of that, and I
think that falls on Congress.
And that is why, when you think about how Congress
approaches this, a point I haven't made at this hearing that is
really important, you know, how--I think this committee has
really changed and evolved from when I first got on it a long
time ago. We are less parochial than we used to be. It is still
a problem. And when Congress--when we want to look at, okay,
how can we help, stop defending every last project that is in
your district and in your state. That really impedes the
ability to be flexible, and it turns it into a fight just to
preserve things for the sake of them being in your district.
And political, I will tell people everyone thinks that is
the key to winning the election. I always love to tell the
story of the A-10. You know, we had to protect the A-10. And I
am not going to get into the argument of whether or not we
really needed it or not. But first, we had to protect it
because of Ron Barber. He was in a swing district, it was going
to be difficult, he was going to get killed if we didn't--we
protected it, he lost.
[Laughter.]
Mr. Smith. Then we had to protect it for Martha McSally.
And we did, and she lost twice.
[Laughter.]
Mr. Smith. I mean, you can get reelected without just
trying to bring home all the money to your district. Honest to
God, you can. So please, let's move past that.
And then the second thing, to defend DoD a little bit, part
of the reason that they don't want to step out and make
difficult, change-oriented decisions is because, in addition to
us being parochial, the other thing we love to do is grandstand
at the expense of the DoD, okay?
If someone makes a decision, makes an innovative decision,
and it doesn't go well, we talk all the time about how DoD
needs to be tolerant of failure. We are not tolerant of
failure, you know, because we want to get--no, I want you to
make sure that we never--this never happens again. You will
rarely hear me ever utter that sentence, you know, because you
just--you can't. And trying to make sure nothing bad ever
happens again leads to all kinds of bad decisions. So I hope we
will think about that, too, as we try to figure out how to work
better with Congress.
The closing comment I think--I want to really amplify what
Congressman McCormick said. Yes, there is problems with
conflicts of interest, there is problems with Congress, there
is problems with prime contractors. But at the end of the day,
the biggest impediment to what we are trying to do here is what
Richard said: resistance to change. Okay? That--if we could
smash any one thing, the resistance to change is what we have
to smash in order to successfully implement the changes that we
have talked about today.
But again, I want to thank the chairman very much for this
hearing. This has been incredibly informative, and we will
build on it going forward.
I yield back.
The Chairman. The chair now recognizes the gentleman from
Florida, Mr. Gaetz.
Mr. Gaetz. Thank you, Mr. Chairman, but we are not in
service of those very valuable goals that the chairman and
ranking member just laid out. We are not on a basketball court
uncontested. You know, there are a group of people assembled to
try to figure out how we never change, because that is how they
make money, and that is how they pay their dividends. And so--
and by the way, that is an incredibly sophisticated
infrastructure.
Mr. Ludwig, you said earlier that the next big innovation
will be a software innovation, not a hardware innovation. Just
expand on that briefly.
Mr. Ludwig. Fundamentally, AI is a software problem. I
think there is a lot of talk about chips and advanced chips.
The reality is the chips are good enough that we have today to
do extremely advanced things. Many of the hardware systems that
are involved in making planes and boats are commodity parts.
The little electric motors, little--the little lithium
batteries that exist in planes, these are commodity parts. The
sensors that exist in your smartphone are plenty powerful
enough to be on a missile. The real innovation and the real
discriminator is going to be the software capabilities that
these systems have.
Mr. Gaetz. Yes, so--and Mr. Sankar, you said the principal
takeaway you want us to draw from your appearance today is the
need for competition, more competition across all of these.
Software is certainly not an exception to that, right?
Mr. Sankar. Absolutely. I would even go even further. It is
like software is a unique American strength that we are so good
at as a country that we tend to underestimate how good we are
at it.
Mr. Gaetz. Right, but we can only be that good with the
competition. And here is --
Mr. Sankar. Yes.
Mr. Gaetz. Here is the point I want to draw out. So we go
back to sort of my favorite, you know, punching bag, the F-35.
So in the F-35, we go buy the $100 million commodity, right,
the hardware, and then we put into our decision-making a full-
system performance contract for Lockheed. So now, if anything
goes wrong with anything software-wise, the point of entry to
be able to solve that problem is the entity that created the
problem and wants you to buy 10 more $100 million paperweights.
So, you know, I guess I would ask, do you believe it is
wise, in this era of great software innovation, for us to ever
again on any platform give a defense company a full-system
performance contract, Mr. Ludwig?
Mr. Ludwig. I strongly feel that Congress should encourage
that hardware and software are competed and procured
separately, and this allows strong software companies to put
their best foot forward and provide their best innovation, and
strong hardware companies can put forward their best
innovation.
Mr. Gaetz. And now, by force of law, we have no ability to
do that. If one of you showed up tomorrow with a widget to make
the software work better on the F-35, you would have no point
of access to be able to present that solution because of the
way we have chosen to give the five legacy defense contractors
these type of full-system performance things.
So, like, while I think--I really think the chairman,
Chairman Wittman, a bunch of our team members on Armed Services
did hard work to try to limit our purchase of the $100 million
paperweight, at the same time what we didn't do was unlock the
competition on the existing commodity. And if we do that, I
think that your goals would be more easily achieved.
So that is all I have, Mr. Chairman, I yield back.
The Chairman. The gentleman from California, Mr. Khanna.
Mr. Khanna. Thank you, Mr. Chairman.
I will be parochial in saying whichever of you two
gentleman are chair next year, I hope this can be a tradition
in coming to my district once a year. I think it is a helpful
thing you have started. And I have two questions.
One, do you have any recommendations and--of people who
understand the industry technology, but also would be effective
in the Department of Defense, who would be good people for
either--the new President to appoint? Does anyone have
recommendations?
Mr. Sankar. I have maybe a broader observation, not
specific names.
One of the things, I have worked with the Israeli
Government since 2014, and it is a country that thinks very
highly of its own technical capabilities, correctly. So when
they mobilized 360,000 reservists after October 7, those
reservists, now with two decades of industry experience, were
horrified at the state of technology in the IDF [Israel Defense
Forces].
That is an implicit self-critique of their younger, 20-
year-old selves that they had all the IQ [intelligence
quotient], they had all the skills, they didn't have the
experience. We have that experience in this country in spades.
And creating opportunities for part-time service members,
leveraging space forces, part-time authorities, creating
opportunities for America's unique tech industry to be able to
contribute, provide advice, guide our senior military leaders
would be uniquely valuable.
And often the most important advice is what not to do,
because that person has already screwed it up in the private
sector and knows that this place, this isn't going to work,
here is an easy way to save $10 billion, and I think it is a
missing natural resource that we are not leveraging. It is
maybe the new equivalent of the dollar-a-year man that we used
to great effect to win World War II.
Mr. Khanna. Mr. Tseng?
Mr. Tseng. So I do have names I could follow up with you
afterwards, but it is more important from, like, who you are
looking for. And I think it is about--you are looking for
people who are working to solve these problems every day, who
have actually solved them in terms of scaling, who know the
problems inside and out.
One of my major challenges, one of the complaints that I
have about the leadership in DoD, they are not actually solving
the problem. They are not on our side of the table, trying to
figure out how to scale--how to how to get, you know, a
company's product into the warfighter's hands every single day.
And so, when I talk to them about the problems, they visualize
them as very abstract. Or if you talk to industry, I think what
you have seen here is very, very tangible--like, every single
person at this table has felt and experienced those problems.
So I claim you want someone, if it relates to acquisition
or scaling innovation in the Defense Department, someone who
has actually gone from zero to a billion in the Defense
Department. And if you look at the list of the names who have
actually been there and done that and led it, there is probably
three on the table.
Mr. Khanna. I would just encourage you to, given that there
is going to be a transition, to be specific. I mean, my view
is, for example, Ash Carter was very, very forward-looking in
that, and made a huge difference. And whether it is at the
Secretary of Defense level or below, if there are specific
names or people you think would be good, giving them to the--
either the chairman or the ranking member, I think, would be
helpful.
And my second question is more broadly I--you know, in
Silicon Valley one of the challenges is a lot of the--
candidly--the recruiting happens at the top 10, 15
``universities.'' And I have always said that DoD could be one
of the places that young people get to go for technology,
talent, where it is not just the top 15 universities, where
they are contributing, getting training, and then can go into
the private sector. How are you seeing that pipeline of young
folks in technology, AI, other places?
And could DoD be one of the first hires for some of those
individuals?
Mr. Valentine. Absolutely, Congressman. I think one of the
most interesting programs that I have seen recently--in another
life I also serve on the Software Defined Warfare Commission
with the Atlantic Council. And one of my fellow commissioners
is the president of Purdue University, and he introduced me to
this new organization called DCTC [Defense Civilian Training
Corps], or the Defense Civilian Technology Corps. So it is like
a ROTC [Reserve Officers' Training Corps] analogy, but for the
civilian portions of the government.
I thought that was fabulous, and so I did some research on
that. I think that is a great program and should be continued.
Mr. Khanna. Terrific. Thank you.
The Chairman. Mr. Gimenez.
Mr. Gimenez. Thank you. And some of the things that were--
that struck me was that there is a lot of stuff that we sent to
Ukraine that they don't use. Is that right?
Do you think that is done on purpose, the stuff that we
know that doesn't work anymore that we are sending over there,
or is it stuff that they asked for that now they figured out
doesn't work? What? Which is it?
Mr. Tseng. I don't think there is--I hope there is not
malintent. I don't think there is malintent. I think it is
people.
When you give something, there is no better customer than a
warfighter using your product. You will find out really fast if
they are going to--that is how we measured, like, our success,
actually, was how often would a warfighter take our product
over someone else's product onto the battlefield? That was the
measure of success. It wasn't--because at the end of the day
they are making a choice. You can only carry so much with you,
and they have to make a very deliberate choice of what they are
bringing. And so they have made their deliberate choice by not
bringing a vast amount of U.S. equipment that is sitting over
there.
Mr. Gimenez. But again, the question again--you kind of
dodged the question, okay?
There is stuff over there. Did they ask for it, or did we
just send it to them?
And if we are sending it to them without them asking for
it, are we sending them stuff that we know doesn't work and we
are just getting rid of it because we know we don't need it
anymore over here?
Mr. Tseng. It is probably a combination of them both.
Mr. Gimenez. That is troubling.
Mr. Tseng. Yes.
Mr. Gimenez. Because then you have this figure of billions
of dollars of American aid that are going to Ukraine that
people are going bananas about, and most--a lot of it, I don't
know how much--we have got to dig into that one--is actually
just stockpiled somewhere because it is useless to the
Ukrainian and to the war effort.
And again, if I were over here, maybe I would want to get
rid of some of that stuff that I know doesn't work, all right,
and send it over to them and let Mikey eat it, okay, because I
don't really want it. So that is something--that is another
avenue that I would like to pursue, but not in this--right
here.
Is it possible with AI and automation that we can turn,
let's say, a fourth-generation fighter into a fifth-generation
fighter, into a sixth-generation fighter, giving them more
capability because you take the pilot out?
Mr. Tseng. Yes.
Mr. Gimenez. Boy --
Mr. Tseng. One hundred percent.
Mr. Gimenez. --on that one.
[Laughter.]
Mr. Tseng. Yes, yes. No, I think about that a lot. There is
4,000 F-16s in the world, right? I tell people there is already
a long laundry list of potential CCAs that we are already
sitting on. But --
Mr. Gimenez. Which is exactly what I want to get to,
because --
Mr. Tseng. Yes.
Mr. Gimenez. --they already are there.
Mr. Tseng. Yes.
Mr. Gimenez. Instead of buying fifth and sixth-generation
stuff --
Mr. Tseng. Right.
Mr. Gimenez. --we already have fourth-generation stuff that
is sitting there. Can we increase the capabilities by
automating them, getting rid of the pilot --
Mr. Tseng. You --
Mr. Gimenez. --getting rid of the life support systems, all
that, and also enhancing the capability of that aircraft?
Because, probably, the aircraft capabilities are hindered by
the human body, they can only take so many Gs, et cetera,
right? AI can take, I don't know how many Gs, you know,
whatever.
Mr. Tseng. You could absolutely retrofit all of those
aircraft. That is--I mean, we--the U.S. Air Force retrofitted
aircraft for the purposes of the DARPA program that Shield AI
was a part of. And so you could do it at a greater scale. There
are pros and cons to doing it.
Mr. Ludwig. I would just add that, absolutely, the most
fragile component of an aircraft at this point is the human
body. The hardware capabilities definitely exceed what is
possible with a human pilot, and I am optimistic that very
advanced AI technologies really could unlock a lot of new
capabilities.
Mr. Gimenez. Yes. Go ahead.
Mr. Sankar. If I could add one point to the point of
competition here, it is like when we consider the next
generation coming, we have already sunset the legacy without
making the next generation prove its worth.
You know, as we think about funding counterfactually, the
F-35, why aren't we encouraging the legacy platforms to
innovate and compete and be relevant in the next fight, make
the F-35 earn its place, rather than bestowing its future and
closing out on all the systems and capabilities that we have
today?
Mr. Gimenez. That is interesting, because we--one of the
problems that we have is the production, right, of these
aircraft, whereas the Chinese right now are able to outproduce
us. Maybe that gap can be, you know, closed by thinking about
it in a different way. And so that I find that to be very
interesting. I would like to get more into it, you know, when
we have some more time.
Finally, on the issue of promotions, I agree with some
folks, I guess, up here that the way that we promote and we
move our personnel around all over the place every three years
because, you know, that has got to be three years, you have got
to move on, is insane. It is actually insane. I was mayor of
Miami-Dade County. I wouldn't say, well, you have been the
director of solid waste for three years. I got to move you now
to animal services. No, okay? You have been a great director of
solid waste, you are going to stay in solid waste, thank you
very much.
But I guess, you know, with them, it is, well, I want to
get to four stars. And I will give you an example. I have a
really, really good commander of Southern Command who is
respected tremendously in South America and Central Area.
Tremendously, okay? They love her. She has to move. Why? Well,
her time is up, right? She can't get another star, all right,
but she has to move. Okay? And so, you know, I think that that
is something else we need to break.
I understand the need to move around and to understand the
entire force and all that. But sometimes, when somebody is
critical, maybe just instead of being, well, that only calls
for a one-star, well, maybe I--okay, now it is a two-star, all
right? Because they have done such a good job. But it stays
there. And so that is just an editorial comment on my side.
Do you guys have anything to comment on that?
Mr. Ludwig. At a high level I absolutely agree. I think it
is important that deep expertise is--it takes a long time to
build. And certainly, we see sometimes that an individual who
has developed deep expertise is then moved on to something
else, and that expertise vacuum is a real problem.
Mr. Tseng. I also completely agree with your comments. It
is also about accountability, right? If you have a program
manager who takes over a program that seems to be failing in
the first year or even the first two weeks, right, how do you
hold that person accountable? You need someone there for the
duration to--you know, them hitting milestones, very clear
milestones along these programs, where they can be held
accountable.
Mr. Gimenez. Okay, thank you.
I yield my time.
The Chairman. Ms. Jacobs of California.
Ms. Jacobs. Thank you. In order to give my friend over here
who represents the F-35 factory a break, I am going to turn us
from F-35 legacy systems to shipbuilding legacy systems.
[Laughter.]
Ms. Jacobs. And I represent San Diego, so I am doing this
to myself, Mark.
Mr. Jenkins, I imagine you on the panel have the most
experience on this, but of course any of you are welcome to
weigh in. Obviously, shipbuilding has been another area where
it has been incredibly hard to modernize and to figure out how
we do things differently. And, you know, in recent years we
have seen success in industries like drones and other areas.
But given the historical challenges and the current need for
shipbuilders to keep pace, what specific strategies or reforms
could you help us think through in shipbuilding or the
shipbuilding industry to embrace the same level of innovation
that we have seen in other areas?
Mr. Jenkins. So thank you for the question. It is a big
question.
Shipbuilding is a complex piece. I think, as earlier--with
the requirements, and having narrower requirements would lead
to more efficient, smaller ships, probably. They are more
effective and they will be dynamically changing. I think
hypersonics have changed the face of naval warfare forever. You
are not going to deploy large ships forward-deployed, they will
be standing back, out of the range. You will transfer to a UUV
[unmanned underwater vehicle] or a submarine world with long-
range fire. So that will be supported by a large fleet of
unmanned systems--air, surface, underwater--to give you the
ISAR [Inverse Synthetic Aperature Radar] targeting information
for your services to hit.
So it is a rapidly evolving piece. We still need ships. It
is not a replacement for ships. You are still always going to
need ships. They are incredibly capable when needed, but we
can't build them fast enough. Even if you had all the money in
the world, with our shipbuilding capacity you could not produce
them faster and better and quicker and better ships. So we have
what we have. Making the best use of the ships we have, but
augmenting them with new devices is going to be the best way to
get more data to support our mission.
Ms. Jacobs. Did anyone else want to add?
Mr. Tseng. I would have one more thing to add on
shipbuilding.
We power HD Hyundai's shipbuilding in Korea. They are one
of the largest. Secretary Del Toro went over there, saw what
they were doing, how automated and wonderful the software was,
and said, ``Why can't we have this in the U.S.?'' And really,
you can. I mean, it is a U.S. company that is powering this.
In my own interactions with the shipbuilding industrial
base, I have been told numerous times, ``Look, if I get more
efficient, I am just going to have to give this profit back in
a recompete,'' and so I think there is a role of thinking
through the complicated incentives that actually drive people
to want to go faster, to make it in their self-interest to
actually serve the National Defense Strategy.
Ms. Jacobs. Mr. Chairman, maybe this can be a good Alabama-
California shipbuilding collaboration.
I also wanted to ask you, I think that we have--we have
foot-stomped a lot, everyone here, that it is a culture issue
at DoD that we are dealing with, right? And some of that
culture is derived from leadership, as Mr. Gimenez, as Mr.
Khanna both have sort of talked about. But obviously, we know
that leadership is only one piece of culture and cultural
resistance. And as folks who, you know, run or work in big,
private-sector companies, do you have any other specific steps
you would advocate for us to think through in terms of, you
know, changing this culture or the entrenched mindset?
So, you know, for instance, examples of accountability
mechanisms, leadership evaluations, innovation, performance
reviews, like, how would you--if you were going to have to sort
of try and, well, for lack of a better metaphor, move the
aircraft carrier that is the DoD into this better, you know,
direction, in addition to leadership, are there other specific
things we should be thinking about?
Mr. Tseng. I just think a lot about incentive structure,
and that is what private industry will focus a lot on, right,
is do you have the right incentives that are going to drive the
right behaviors? Do you have the right accountability that is
going to drive the right behaviors, as well? But those two go
hand in hand.
But you can incentivize people to do incredible--there are
incredible people, as you all know, in the Defense Department.
Sometimes they are stuck under a rock. But if you have the
right incentive structure in place, they can move mountains,
and will move mountains.
Mr. Sankar. I would add one comment. I would second Ranking
Member Smith's comments that, really, we need to create a
culture that fosters risk acceptance. When we are doing new
venture formation, you are lucky if 1 out of 10 of these things
really work, and you are betting on the trend, the ability to
learn, you know? So thinking about how we encourage and accept
the risk in the portfolios that we are going after here is, I
think, very important.
I hear, almost as like a shibboleth, people, general
officers will tell me, well, we have proven that, you know, we
need the congressional oversight because we have proven that we
are not good at spending taxpayer money.
Ms. Jacobs. That is true.
[Laughter.]
Mr. Sankar. But it has almost become--it becomes self-
defeating, you know? It is like by creating processes that make
sure that nothing goes wrong, you also create processes that
make sure nothing can go right, and it constrains you to
mediocrity.
And so I would gladly accept more failure if it meant that
we had more catastrophic success, as well.
Ms. Jacobs. Yes. And my last question--again, for Mr.
Jenkins--you know, Saildrone has been one of the big success
stories of DIU. How can we do more of that? How can we have
more success stories, right? Because for every one of you there
is a number of companies that weren't able to get the funding
they needed or, you know, couldn't get through the bureaucracy,
even with DIU. So do you have suggestions for us on how to make
DIU easier to work with, or anything like that?
Mr. Jenkins. So I think it is a good time to point out
that, even though we are hailed as one of the best success
stories, we were one of the first projects funded by Ash Carter
back in 2016. It had incredible impact to the Navy and
operational field. We still do not have a program of record, so
the system has not been able to get itself into a way that says
this is a--not only just promising, this is a system that does
what we want, it is cheaper, it is better, it is faster. They
can't pay for it.
And what we are seeing is a whole plethora of Band-Aids
of--whether it is SBIRs or APFIT [Accelerate the Procurement
and Fielding of Innovative Technologies] or Replicator, they
are all just congressional plus-ups as Band-Aids to try and
give them more things. And what we need to do is say, no, no,
there is lots of amazing R&D innovation. How do you take that
and put it into operations?
So as you are saying, it is incentivizing the folks who
control the COCOMs budget, the fleet budgets. They have to be
incentivized to take risks. Otherwise, the system will never
reform itself. So lots of ways to do that, but I think DIU--off
your question, but DIU has been a great initiative. It has got
us to where we are. It is fantastic, huge support, I thank them
a lot. But there is almost no stepping stone to go into
operation, so we have to create that vehicle to do that.
Again, DIU has done it very successfully, done the--kind of
the project-based requirements. Rather than setting strict
requirements it said, we have this problem, give me air
superiority, I don't care whether it is a plane or a missile or
a bomb, and that has been very quick to get commercial minds on
the case, which is what drives innovation.
Ms. Jacobs. Got it. Thank you.
The Chairman. The gentlelady from South Carolina, Ms. Mace.
Ms. Mace. Thank you, Mr. Chairman.
And I believe, Mr. Valentine, you mentioned it sort of in
passing earlier, and I would like to maybe have a slightly more
in-depth conversation. Talk to us about chips.
Mr. Valentine. Yes, Congresswoman. I think it is pretty
well known that we have a deficit of the ability--most of our
chips are made in Taiwan. They might have been designed in the
United States. We send those designs out. They are actually
etched, lithography happens there. Then they are delivered
back. So that creates a vulnerability in getting those advanced
chips that we may need at any one time.
So I personally think that whatever we can do to start
onshoring that capability so that we have access to those chips
in time of need is of strategic importance.
Ms. Mace. Anyone else want to chime in on the issue of
chips and how we get them here, how do we solve the problem?
And then I guess, two, the other question I have, I would
like to hear from everyone who is here today about, you know,
the advances in AI and technology, the amount of energy that is
needed and going to be needed in the near future to keep up
with the advances in how fast things are going. The data
centers. Is it nuclear? What is it? How do we handle what is
going to come in terms of--in order to keep up with the
technology we have got to have the energy capability, the
energy infrastructure. We have to have the data centers to be
able to keep up. You know, how far behind are we on this? How
do we solve this problem? How do we keep up with what is
happening in this space?
Mr. Sankar, you can start.
Mr. Sankar. Thank you. I think this is an opportunity to
really organize a whole-of-government effort.
Like, what would it take for us to have 10 times the amount
of energy production? We have been roughly flat for the better
part of three decades or more now, and I think it is going to
pose challenges on the grid. The answer is we need all of every
type. You know, I don't think it is going to be a single
modality. But the prosperity, the economic prosperity of our
nation here so profoundly depends on this singular starting
point and the--kind of the value generation supply chain of
having enough power to continue to create the advances and
maintain our lead as the dominant country with AI.
Ms. Mace. Does anyone else want to chime in?
All right, that is all I had, Mr. Chairman, I yield back.
The Chairman. Mr. Veasey, Texas.
Mr. Veasey. Thank you, Mr. Chairman. I just have one
question for Mr. Jenkins.
You said something very briefly in your testimony earlier
about the Chinese, and I guess the--a vehicle that they have
that is much superior, from certain technological standpoints,
than Tesla. And I wanted you to touch on how that should be
worrisome, from a military standpoint and all the things that
we have talked about here today, because we are still having
some huge philosophical disagreements on, you know, EVs
[electric vehicles] and what role they are going to play in
just everyday American society.
Mr. Jenkins. Yes, I think that wasn't my testimony. It
might have been one of --
Mr. Veasey. Oh, maybe it was--okay, I am sorry, I
apologize. Yes.
Mr. Ludwig. So I think the important thing to understand is
that there is an enormous amount of innovation happening in
China. The automotive industry there is incredibly competitive.
There is over 140 Chinese car companies. And of those 140, I
would say probably 20 of them are, I would say, best in breed.
That would be fully competitive globally.
Importantly, the rate of software updates that go to their
vehicles is much faster than what you would typically be used
to in most types of vehicles on the road. And the pace of
innovation is so fast that leaders in the auto industry, they
are now making it a point to travel to China on the order of
every three months, just to see the latest software updates
that are going to these vehicles. And this really underlines
the importance of enabling rapid software updates for all of
our hardware systems, because hardware can become obsolete very
quickly without up-to-the-minute software.
The Chairman. Mrs. Kiggans.
Mrs. Kiggans. I just was curious about a little more
clarification about some of the Ukraine aid that Congress has
approved, and you mentioned that you got zero of those dollars.
So are you benefitting at all?
Because I know you talked a lot about all of your
involvement, some more than others. But, you know, is it
worthwhile for us to continue to invest? I want to make sure
our dollars are--it is easier to get more of Congress to
approve this type of aid if we can guarantee that our dollars
are benefitting our defense industry. So it would be helpful to
know--or maybe to not, if you can shed light on that--if that
aid that we are approving is somehow helping you all.
Mr. Valentine. Congresswoman, again, I can't directly trace
a dollar that has currently bought one of our drones to send
them over. I know the U.S. Department of State has purchased
several that they have delivered to both members of the
Ministry of Interior and other organizations within Ukraine, to
include the office of the Prosecutor General. But to be honest,
I don't know where those dollars are coming from.
I will tell you, though, the funding is making a
difference, and I think it is instrumental because it helps
Ukraine, it helps America and our industrial base and, quite
frankly, increases our competitiveness, which hurts our
competitors. So from those three perspectives, I think it is a
no-brainer.
Mr. Tseng. I would just add I think it is a good thing. I
think it incentivizes companies to go out there and solve the
hard problems, knowing that there is a--knowing that there is
opportunity, should you solve those hard problems, it gets a
lot of companies to go out and figure out how to solve them.
Mrs. Kiggans. And it sounds like--and I think you even used
the words--it was a laboratory or test bed for you all. So that
is good, because we know the battlefield is changing. We know
it has changed from Vietnam to when we fought in the Middle
East to now, what is going on with Russia and Ukraine. And we
know that the battlefield for probably Southeast Asia is going
to look different, as well. So you have probably made a lot of
progress in that European domain space, and that--what that
Russia-Ukraine battle looks like.
But are you able to make any progress in Southeast Asia? We
would probably predict that that is where, potentially, our
potential next fight would be at. So is there anything going on
that is enabling you to make the kind of progress that you have
made in Ukraine, and apply that to Southeast Asia?
Mr. Tseng. Yes. The principal difference in Ukraine is
actually something everybody expects in a China-Taiwan
conflict, as well, and that is just the prevalence of
electronic warfare, right?
The Chinese see that the Russians are being very successful
with electronic warfare, jamming GPS, jamming communications.
And the Ukrainians, we see that--the Ukrainians are also very
successful jamming GPS and jamming communications. And so I
think it is very reasonable--it is reasonable to expect
everybody in the DoD knows, and China has tested this in the
Strait where they are jamming GPS and communications. So
there--the point of a battle lab of what is going on in
Ukraine, is it going to have relevance in China-Taiwan?
Absolutely.
There are other subtle differences, right, okay, it is a
principally maritime environment. The geographies or the
distances are much longer because it is a maritime environment,
because it is an island. But I think that those core principles
of, hey, you are going to be operating without GPS and
communications, those are enduring.
Mrs. Kiggans. That is good to hear. And then I guess those
lessons learned that you all have learned--because we obviously
don't have U.S. troops on the ground, thankfully, in that
conflict--so are you able to communicate, and is the military
listening to those lessons learned, and be able to hopefully
apply that to--I know you have talked a little bit about how we
are not practicing how we would, you know, conduct warfare. So
I just want to make sure those--the things that you all have
learned as civilian companies are getting translated to the
military.
Mr. Tseng. They are, and I--well, there are pockets of
success. That is what I will say. It is not--so for me to say
broadly, yes, everything gets disseminated down the chain of
command, I would think, would be--or to every operational
command is absolutely not true. But there are pockets of
success where commands take those learnings and translate them
down into action.
Mrs. Kiggans. Well, please be a loud, squeaky wheel, and we
will try to help.
Thank you. I yield back.
The Chairman. The gentleman from Georgia, Mr. McCormick.
Dr. McCormick. Thank you, Mr. Chair.
Interesting, Mr. Gimenez, I thought you brought up an
interesting point about retrofitting aircraft. My personal
opinion, I don't think it is worth it, and here is why,
because, like, take an F-16, who has a great roll ratio, one of
the fastest, one of the best G ratios. The problem is that it
is still designed around a human to begin with, and so you are
never going to be able to upgrade the capability of stressing
the airframe beyond the human being, which we designed it
around.
And so--and quite frankly, the biggest loser in all of this
is really the good movies we are going to make.
[Laughter.]
Dr. McCormick. Because when the dogfighting is over--and
those days are actually gone already, I don't even know why we
bother doing dogfighting, because, quite frankly, it is back in
the Yeager days. Yeager used to always say that the first pilot
who sees the other pilot wins, because you put yourself in
that--those days have been gone for a long, long, long time
now. It is the best person who has the best radar and the
weapon that can fire the furthest way wins because you are
going to fire before you ever see them and you turn around.
That is our dogfight now. It is ridiculous that we even have G-
rated aircraft that--I mean, other than to evade. And we know
that our evasion capabilities aren't going to exceed whatever
some little tiny missile is for Gs.
So we are kind of, once again, thinking futuristically.
Retrograding doesn't really help us quite--in my opinion. We
need systems now that are smaller, more maneuverable, require
less fuel. These systems are based way too large, and we need
to scale down. That is my personal opinion on that as a kind
of--a pseudo-expert, if you will.
Ukraine. You know, one of the things I thought was
interesting, everybody gets really upset in America about now
wasting money in Ukraine, and how inefficient we are and we are
sending the wrong weapon systems. It pales to what we did in
Afghanistan. We did not learn a lot of lessons in Afghanistan.
We should have. But one of the things is we sent so much money
and so many things that they could not use over there
constantly because of our contracting--and this is recent, by
the way--that it was criminal when I was over there. Literally,
we had parts to show up for five tons that were from Humvees on
purpose. And I know it was on purpose, and we did nothing about
it in our accountability. And now, all of a sudden, we are
going to be accountable in Ukraine, where we are 20 times at
least more accountable in Ukraine than we ever were in
Afghanistan, when we never talked about it. I think it is
hypocritical.
So I hope we do get our stuff together when it comes to
Ukraine, because the American public is watching. We are
accountable to them. And so--and then so those are two topics
somebody else already--that had very strong opinions on, I just
had to state.
And then the final one Ms. Mace brought up I thought was
really interesting, when it comes to what TSMC [Taiwan
Semiconductor Manufacturing Company] does for us and the AI
production. We have the CHIPS [Creating Helpful Incentives to
Produce Semiconductors] Act. But in my opinion--I would like to
hear your opinion as a community --I don't think the CHIPS Act
really brings enough home when it comes to the most advanced
chips, the AI chips, because the CHIPS Act addresses chips that
we use for our cars, you know, cars that drive themselves,
stuff like that, that is great. But the real AI, the leading-
edge technologies, we are not bringing home. And we do have an
adversary who is saying, ``We are going to take Taiwan,'' which
means that they are going to control basically 100 percent of
our AI production in the future. I don't think we are talking
about it. I think we thought the CHIPS Act solved it. I don't
think we are having a realistic conversation in Congress about
that.
Do you agree or do you disagree with my assessment of not
bringing home enough--quickly enough, given the threat to
taking out 100 percent of our chip production on the leading
edge?
Mr. Valentine. Congressman, I admit I am not an expert on
the CHIPS Act, but I agree with your thesis that we need to
bring home the chips, and do it fast.
Mr. Tseng. I would just add to that. I think I applaud
Congress for pulling the CHIPS Act together. I think the--you
know, there are things money can't buy, implementation. And so
the Japanese started building their TSMC fab a year after us,
and they are already done. You know, where is the organizing,
whole-of-government effort to pull this together as a cross-
functional effort to make sure that we build this thing as
quickly as possible, we solve all the problems?
There is going to be so many problems between here or
there. It is not as simple as just cutting a check. So we have
got to organize around it, as this nation is so capable of
doing.
Dr. McCormick. And just--I think we need to do the same
thing with our production of medications, because I think there
is too much quality control we don't have that is over there
that we need to bring back home because we are getting bad
drugs, poisonous drugs, ineffective drugs, and we don't do any
sort of investigation like we do on our own homegrown products,
and I think it is absolutely dangerous to our future.
Thank you. With that I yield.
The Chairman. I really appreciate you all making yourselves
available. This has been very helpful, and I thank you very
much.
We are adjourned.
[Whereupon, at 11:39 a.m. PST, the committee was
adjourned.]
=======================================================================
A P P E N D I X
September 16, 2024
=======================================================================
PREPARED STATEMENTS SUBMITTED FOR THE RECORD
September 16, 2024
=======================================================================
[GRAPHIC(S) NOT AVAILABLE IN TIFF FORMAT]
[all]