American Global Competitiveness at 250: Legislative Proposals to Secure U.S. Technology Leadership.
Legend
Transcript markers
- text Sentence being played
- text Search result
- text Intervention cited by the link you opened
Click a sentence to play the video from that moment. Select text to quote it with the speaker, timestamp and link.
More options
House Energy and Commerce Subcommittee on Commerce, Manufacturing, and Trade U.S. House of Representatives
Follow its sittingsKathy Castor
It would give companies an impossible mandate. It tells any American company doing business in Europe, our largest trading partner, that it must break European law to satisfy this Congress, or follow European law and be punished here at home. This bill does not make American companies more competitive. It shields polluters and corporate bad actors from environmental accountability, while leaving responsible businesses trapped between conflicting laws. We should instead refocus on competitiveness through innovation instead of economic self-sabotage.
We already know how to do this. When this committee passed CHIPS and Science, we bet on American science and American workers. And that bet is paying off in plants and jobs across the country. That is what true competitiveness looks like. So I look forward to hearing from our witnesses today and furthering this discussion.
Thank you, and I yield back.
Jay Obernolte
Gentlelady yields back. Appreciate, and I'll recognize Mr. Obernolte for actually pinch-hitting for our good chairman, and I'll recognize him for five minutes for his opening statement.
Thank you to all of our witnesses for being here today. For decades, the United States has been the world leader in technological advancements. Thank you, Mr. Chairman. It's an honor to give this opening statement on behalf of Chairman Guthrie. Good afternoon.
American innovation in AI, semiconductors, quantum, and other critical industries has set the foundation for our digital economy. To maintain U.S. global leadership, we must create the right conditions that enable U.S. innovators to flourish on the world stage. We cannot allow our adversaries, such as China, to overtake our position through aggressive state-driven industrial strategies. Today's legislative hearing takes an important step in evaluating policies necessary to position America's technological leadership and global competitiveness. One element of this is addressing foreign regulatory regimes, such as the EU Corporate Sustainability Due Diligence Directive, which reaches far beyond the EU's jurisdiction and would impose significant compliance costs on American companies, diverting resources away from investment, innovation, and job creation here at home.
To remain a global powerhouse, we must bolster our supply chains and support the adoption of critical and emerging technologies, which have the potential to define the future of American prosperity. Another key focus of this committee is the semiconductor industry, which is foundational to nearly every modern technology and a core national security priority. As demand grows and global competition intensifies, it is essential that U.S. semiconductor manufacturing capacity continues to expand and outpace foreign competitors. I'm encouraged by the discussion today, and I'm confident that by working together, we can develop policies that keep the United States at the forefront of global innovation and ensure our leadership for generations to come. I yield back.
Speaker 1
Gentleman yields back. We appreciate, and now recognize the ranking member of the full committee, Mr. Pallone, for his five minutes and giving his opening statement. But before I do that, I want to congratulate all the committee members, the chairman and the ranking member and all the members for that bipartisan win yesterday on the floor. We passed the KIDS Act, and it was a, I'm tell you what, what a victory. Now we got to get it across the finish line and get it to the president's desk.
But let's keep it going, guys, because we're doing really good work for the American people. So with that, I will recognize the ranking member of the full committee, and I appreciate all his work. Thank you.
Frank Pallone
Our ranking member is Castor for all the work that she's spent on this over the years. The two of you really did a lot of the work on this and I do want to recognize that as well as well as Chairman Guthrie. But let me get back to this issue. So when we talk about the issue before us today, I just want to stress that American ingenuity and inventiveness are things that we can all celebrate. And over the course of our 250-year history, America has been the birthplace of countless innovations from telephones to airplanes to the technology that underlies the internet.
And all of us want to see that success continue over the next 250 years. But there's a growing there's growing evidence that China is pulling ahead in research and technological development. And while China advances an integrated industrial strategy, the Trump administration unfortunately is dismantling ours. The Trump administration illegally froze more than 7,800 federal research grants. The cost of these research cuts is enormous.
Studies estimate as much as a trillion dollars in GDP. And top talent is leaving the country or choosing not to come here in the first place, handing a generational advantage to our competitors. The Trump administration has also repeatedly tried to shut down the manufacturing institutes that train our workforce and develop the advanced manufacturing technologies of the future. And just this May, the administration proposed rule changes for federal grants that politicize science and threaten the independence of the scientific research that has made America research the envy of the world. We can't compete with China and others while gutting the very institutions designed to make competition possible.
And Democrats have offered a different path. Through investments like the CHIPS and Science Act, Democrats fought for the future of American innovation with historic investments to modernize critical domestic sectors and create jobs for Americans. The CHIPS and Science Act invested nearly 53 billion dollars in the domestic semiconductor supply chain and every public dollar has catalyzed five to seven dollars of private investment. Since enactment, private companies have created 58,000 jobs across 28 states and committed more than 450 billion dollars in investment. And the CHIPS and Science Act invested in America, our workers, and our industries.
Many of the bills before us today address important sectors and technologies for American economic leadership. For example, the CHIP EQUIP Act will ensure that CHIPS and Science Act dollars are not used to bolster our adversary's capacity. Other bills and discussion drafts in this hearing address important issues, including robotics, immersive technologies, biosecurity, biomanufacturing, and supply chains for quantum technologies, automobiles, and memory chips. And I look forward to working with my colleagues across the aisle on these bills. I do have concerns about the Protect USA Act, which would force American companies to choose between violating the act, which exempts companies from foreign regulations related to sustainability and social impacts, or facing penalties in the European Union.
This would push our allies away at the very moment we need them in our corner to coordinate a response to China's sustained strategy of hollowing out critical economic sectors. And it would and in my opinion, it would cut American businesses off from global competition, which will make them weaker in the long run, not stronger. So ultimately, American leadership in the next generation of technologies requires investment in research and talent, not the cuts we've seen from the Trump administration. And with that, I thank you and yield back, Mr. Chairman.
Speaker 1
Thank you, I appreciate it very much. And we'll go ahead and recognize the witnesses. Our witnesses today are Mr. Jason Fiorillo. He's the Chief Legal Officer for Boston Dynamics. Thank you, sir, for being here.
Next, we have Mr. Neil Chilson, who is head of AI policy at Abundance Institute. Thanks for being here. And next, we have Dr. Jedidiah Isler. I hope I said that right. Chief Science Officer for the Federation of American Scientists.
And next, we have Mr. Marty Durbin, Senior Vice President for Policy at the US Chamber of Commerce.
So I'll recognize Mr. Jason Fiorillo for his five minutes of testimony. You're recognized, sir.
Jason Fiorillo
Chairman Bilirakis, Ranking Member Schakowsky, and distinguished members of the subcommittee, thank you for this opportunity. I'm the Chief Legal Officer at Boston Dynamics, one of the world's leading developers of advanced mobile robots. Previously, I worked at a global semiconductor company. Collectively, I've devoted nearly 20 years to enabling domestic enterprises to service trusted sources of critical technology to the United States, its allies, their commercial and industrial enterprises. I'm here to request your support for measures that advance the domestic development, scaling, and deployment of physical AI and to position the United States to lead in the manufacturing and use of advanced mobile robotics.
Boston Dynamics was founded more than 30 years ago as a spin-off from MIT. For much of our history, we served as a pioneer in the mobile robotics space. Six years ago, our Spot robot became the first commercially available industrial-grade quadruped in the world. At that time, our pioneering humanoid robot Atlas stood alone in its unique capabilities. Our company now has 1,100 employees, and we plan to double both the number of employees and our footprint in the near future.
Our robots are used in manufacturing facilities, semiconductor manufacturing plants, energy production, and by first responders. Our robots save lives and have even taken bullets in the line of duty. They are used by DHS, the military, and federal law enforcement to neutralize improvised explosive devices, provide perimeter security, and facilitate safe entry in dangerous spaces. In such sensitive applications, robots that contain cyber vulnerabilities or backdoors, especially those that are produced by adversarial actors, pose a variety of very serious risks. Inspired by our success, others have followed in hot pursuit.
China has more than 140 humanoid robot manufacturers who have been hyper-accelerated by governmental programs and state sponsorship. China's strategy for advanced robotic technologies mimics the approaches that it has used in other sectors. It markets products at prices only possible by government support or overproduction. Most recently, China's government has issued a directive to its provinces and local governments to deploy at least 10,000 humanoid robots in more than 100 high-value application scenarios, with state-owned enterprises directed to serve as real-world training spaces. The physical AI race is well underway, with strategic implications for U.S. manufacturing, energy, public safety, and national security.
Left unaddressed by policymakers, Chinese suppliers will flood global markets with below-cost, government-subsidized robots and will drown competition. An all-of-government strategic approach is necessary to support the long-term competitiveness of the advanced mobile robotics industry. It is time for a national robotics strategy. We have been advocating for one for the past three years, but this has now become a view expressed by an overwhelming consensus of think tank leaders and trade associations, including A3, AUVSI, SCSP, ITIF, CSIS, the Jamestown Foundation, and AEI. Such a strategy would comprise a broad set of policies designed to support the growth and success of the industry.
The recently introduced National Commission on Robotics Act on today's agenda would kickstart this process. We strongly encourage Congress to promptly pass this bill. While that legislation advances and the commission gets to work, we cannot let time slip by.
In the interim, government should take the following four concrete steps. First, government agencies should adopt and use advanced robotics technologies. Congress should instruct and fund agencies to deploy advanced mobile robots in the interest of government efficiency and economic competitiveness, and to save lives and improve society. Second, Congress should explore tax mechanisms that incentivize the domestic manufacture and adoption of advanced mobile robots, including accelerated depreciation, tax credits, and grants. Third, the government should champion and promote the use of advanced mobile robotics across industrial domains, articulate best practices in close coordination with industry, and encourage the embrace of advanced robotics technologies.
And fourth, the government should stimulate domestic and friend-shored supply chains. We should be thoughtful to address high-tech and low-tech ingredients alike. As the physical manifestation of AI, the robotics industry is developing and evolving at an exponential pace. There is broad consensus in support of a national robotics strategy and an urgent need for the government to act now. Thank you.
Speaker 1
Thank you for your testimony, Mr. Fiorillo. Now we'll recognize Mr. Neil Chilson, again, head of AI policy, Abundance Institute. You're recognized for five minutes, sir.
Neil Chilson
Thank you, Chairman Bilirakis, Ranking Member Schakowsky, and members of the subcommittee. Thank you for the opportunity to testify. I'm Neil Chilson, I'm the head of AI policy at the Abundance Institute, which is a nonprofit that's working to build a policy environment where emerging technologies can thrive and enable widespread human prosperity. American competitiveness relies on three capabilities: the capability to invent new technology, to commercialize it, and to deploy or adopt it. Over the past 250 years, we have become the best in the world at all three, but that lead is neither permanent nor inevitable, and today China is nipping at our heels.
So I'm glad this committee is engaged. These 10 bills are evidence that Congress is serious about keeping our edge. I support these efforts and I look forward to discussing their details during questions. I did want to focus in these remarks on the capability that I think we overlook more often than not: the third one, deployment and adoption. Discovery and commercialization, they get a lot of policy attention.
Basic research and technology transfer policies are often mentioned on the Hill. But a breakthrough that's been invented and commercialized but sits waiting for a government permission slip cannot raise wages, lift productivity, save lives, or strengthen our security. Technology leadership requires more than just inventing the future, it requires putting the future to work. And we have a problem. America is now inventing the future faster than we will allow our builders to use it.
Look at AI. Right now, government decision-makers are holding back our very best AI models, OpenAI's ChatGPT 5.6 and Anthropic's Fable 5, from American businesses, entrepreneurs, and consumers who would put them to work. Other than vague cybersecurity allegations, we don't know why they're being held back. There is no clear standard to meet, no defined process, and no timeline for resolution. These are questions that Congress should address.
At the same time, our states have introduced more than 1,700 AI bills this year, a rising tide of confusion and cost for those who build and use AI. We are in effect tying our own hands. And at this very moment, China is giving its AI technology away. It's releasing open-weight AI models that are cheap, capable, and they're fast becoming the world's default. That's why, among the other bills that we'll discuss today, I support the Open Source AI Leadership Act.
US-built open models matter. They are a vital source of competition, innovation, and even security. They let businesses run AI locally, customize it, and build on it without depending on a single vendor. They put American technology and American values in the hands of users around the world.
But today, China dominates this space, producing the best-performing and most widely used open models on earth. The Open Source AI Leadership Act is a good first step to get American open models back on track, and I strongly support it. Turning from China to Europe, American entrepreneurs should not have to answer to faraway governments they had no say in electing. I've spent 20 years in technology policy and I've watched Europe's regulators reach further and further beyond their borders, pressing their preferences onto American firms or extracting billions when those firms resist. Europe's new corporate due diligence mandate, which the Protect USA Act addresses, is only the latest example.
European laws like the Digital Markets Act clearly target American companies. I support the Protect USA Act as a valuable statement of an important national policy. Opening a market to American companies does not grant a foreign government unlimited authority over how that company operates everywhere else. Congress should extend that principle to other overreaching foreign laws. But we should also be honest that the same thing is happening here inside our borders.
A few large states, California, New York, Illinois, are writing technology rules designed to bind the whole country. Whether the mandate comes from Brussels or Sacramento, one capital should not set the rule for all 50 states. Let me close where I began. America still invents like no one else on earth. The question before this committee is whether we will let ourselves use what we invent, and whether we will keep the right to set our own rules.
Get that correct and America's next 250 years can look a lot like the last. Thank you. I look forward to your questions.
Speaker 1
Thank you for your testimony. Appreciate it. Now I recognize Dr. Isler for her five minutes of testimony. Thank you.
Jedidiah Isler
Thank you, Chairman Bilirakis, Ranking Member Pallone, Ranking Member Schakowsky, for the opportunity and the members of the committee for the opportunity to speak. My name is Dr. Jedidah Isler. I'm an observational astrophysicist by training and Chief Science Officer at the Federation of American Scientists. FAS is a nonpartisan, nonprofit science and technology policy organization dedicated to using science, technology, and innovation to improve the lives of the American people and strengthen the nation. That mission is not new to us.
FAS was founded by scientists who understood through World War II and the dawn of the atomic age that scientific discovery could become world-shaping technology faster than our ability of our democratic institutions to govern. That founding insight remains true today. The bills before this committee cover a wide set of technologies, from robotics, immersive technology, semiconductors, open source and open weight artificial intelligence, quantum engineering, biotechnology, memory chips, and connected and autonomous vehicles. These technologies and the bills under consideration differ in design, scope, and market maturity. But they share a simple premise: American leadership in critical technologies is valuable, vulnerable, and worth protecting.
I agree with that premise, but my core message today is that technology leadership is not self-executing. It's not a subscription service that renews automatically. It's more like a garden that must be cultivated over time. The technologies we want to see, the chips, the AI models, the biofoundries, are the harvest and not the seed. But the harvest depends on the soil.
That soil is technically trained people who animate our basic and user-inspired research. It is standards, testbeds, commercialization pathways, and manufacturing capacity. And it is the federal government capable of moving at the speed of technology. At FAS, we approach technology competitiveness from this science-first perspective because science underlies technology. Technology is what science becomes when knowledge is translated into tools, products, systems, and capabilities.
The United States cannot lead in the technology of the future while neglecting the scientific enterprise and, most critically, the people that produce them. That's why a hearing on American technology competitiveness must also be a hearing about American science, talent, and government capacity. America's scientific leadership has always been a people story. The United States became the world's science and technology superpower precisely because it built institutions capable of training, attracting, and empowering extraordinary talent, including Robert Oppenheimer, Carl Sagan, and Katherine Johnson, just to name three. Generations of students, technologists, engineers, researchers, program managers, entrepreneurs, and skilled workers turned public investment into discovery and discovery into public benefit.
Some of these people were trained in American public schools, community colleges, universities, and national laboratories. Many came from around the world because the United States offered something rare: an opportunity for open inquiry, world-class institutions, and freedom to pursue ambitious ideas. Training more Americans in STEM and attracting the world's best scientists and engineers are not competing strategies. They are mutually reinforcing requirements for American technology leadership. As Representative Castor mentioned, you can't decide one year and then deploy technology and technologists the next.
It takes on average eight years after a bachelor's degree to arrive at technical expertise in many fields. Those folks design, build, and operate systems from quantum labs to robotics companies. When we lose that talent, we lose capacity that cannot be replaced on short notice. In a world of technology leadership, it's more than who sells the final product. It's about who trains the workforce, who owns the equipment, who sets the standards, who controls the supply chain, and who can translate discovery into deployment effectively.
That brings me to the legislation before the committee, which I'm happy to engage with individually during the Q and A. Congress should evaluate each of these bills not only by the technology domain it names, but whether it strengthens the underlying system that allows the United States to lead in that domain. A commission, a study, procurement guidelines can be useful if they're producing actionable recommendations and follow-through. All could be bolstered by adequately resourcing the full R and D development lifecycle. Congress has shown it can act deeply and impactfully in bipartisan ways with CHIPS and Science, ARPA-H, and a variety of other bipartisan legislations.
Our next step is to make those pieces durable, fully implementable, and connected to research and workforce base. Thank you, and I look forward to your questions.
Speaker 1
Thank you so very much. Now, Mr. Durbin, you're recognized for five minutes.
Marty Durbin
Thank you, Chairman Bilirakis, Ranking Member Schakowsky, and members of the subcommittee. Thank you for holding today's hearing on American leadership in emerging technologies and global competitiveness. Technologies like artificial intelligence, robotics, and quantum are not future concepts. They're already changing how businesses operate, how workers do their jobs, how consumers receive services, and how our nation competes. They're helping manufacturers improve productivity, small businesses compete and reach new customers, and they support better healthcare, stronger financial services, more resilient supply chains, and more efficient energy systems.
But other nations, as we've heard, are moving aggressively to dominate these fields. America needs a serious bipartisan strategy that promotes domestic innovation, protects critical infrastructure, and strengthens supply chains. We need sustained federal R&D, strong public-private collaboration, cybersecurity preparedness, and a workforce to meet the moment. As a model, the Energy Act of 2020 was a bipartisan act that provided a comprehensive update to U.S. energy policy that promotes innovation across federal agencies and the private sector. And proposals being considered by this committee today can help keep us on that path.
Our capital markets are the envy of the world, and we have the talent, companies, research institutions, and entrepreneurial culture to lead. But leadership's not guaranteed. We must make the right policy choices now. We must also push back against foreign regulatory overreach. The European Union's Corporate Sustainability Due Diligence Directive, or CS3D, is a good example of the broader challenge that we face.
Now, the Chamber supports responsible business conduct, strong transatlantic cooperation, and practical efforts to address environmental and human rights risks. As written, however, CS3D would impose extraterritorial EU requirements on U.S.-parented companies, their affiliates, and business relationships around the world, including activities with no meaningful connection to the European Union. It would require covered American firms to conduct extensive due diligence across global supply chains, creating legal obligations that directly conflict with federal and state law. It exposes companies to enforcement actions, penalties, and litigation for conduct that occurs in the United States and is entirely lawful here in the United States. It further exposes U.S. companies to liability risk in the EU.
Some nations may restrict the audits, data collection, or supply chain investigations needed to demonstrate compliance with CS3D. This creates an EU compliance trap, as companies will be required to gather information that other foreign jurisdictions restrict. Ironically, this is even at odds with the EU's own blocking statute and anti-coercion instrument, which aim to protect EU companies from foreign legal overreach. But the directive also raises serious governance concerns. U.S. corporate law has long relied on principles of board discretion, materiality, and accountability under domestic law.
CS3D imposes prescriptive obligations tied to sustainability objectives and gives broadly defined stakeholders a voice in challenging company decisions. This shifts business judgments away from boards, shareholders, and U.S. regulators toward a European-focused enforcement model. Now, in addition to CS3D, the EU methane emissions regulation follows a similar pattern, affecting U.S. energy exporters based on their commercial relationships with the European market. This rule creates overlapping compliance obligations, uncertainty for long-term contracts, and new risks for transatlantic energy security. We believe this regulation needs to be delayed and reopened.
Taken together, CS3D and the methane regulation reflect a broader trend. Brussels' extraterritorial approach risks placing American companies at a competitive disadvantage. And if left unaddressed, this trend can weaken U.S. competitiveness. It's also a warning for emerging technology policy. The U.S. should not import overly prescriptive foreign regulatory models for AI, robotics, quantum, data governance, digital markets, or sustainability.
Nor should we allow foreign governments to become the default rule-setters for American innovators. America can lead the next era of technology, but we must choose to lead with policies that encourage innovation, protect consumers, strengthen security, empower workers, and allow businesses of every size to compete. We stand ready to help accomplish these goals. Thank you.
Speaker 1
Thank you very much. Excellent testimony from all of you. And now I'll recognize myself for five minutes to begin questioning. Yeah, we're good. Okay.
That's okay. Mr. Chilson, biotechnology is advancing rapidly, as you know, creating tremendous opportunities but also new risks. In your view, what approach does the Biosecurity Modernization and Innovation Act take to promote innovation in the field and address the security risks of synthetic gene synthesis?
Neil Chilson
Thank you for the question, Chairman. So the Biosecurity Modernization and Innovation Act tries to create a national security floor for synthetic gene analysis while preserving space for responsible biotechnology innovation. So it addresses a real market failure here. You know, a person seeking to build a dangerous biological sequence can impose risks on millions of people. There's also the concern that bad actors could try to structure the orders that they have in a way that splits them up among many different providers, and the bill tries to address that through a split order risk.
And it replaces a pretty fragmented voluntary framework right now with something that's clearer. So it has these pro-innovation elements that I think are important and deal with a real challenge. I will say, you know, the implementation here matters enormously. You know, I've studied regulatory frameworks for 20 years at this point, and industry can use these types of frameworks to privilege themselves in some cases or to push out competitors in small areas. And my biggest concern would be that this become a sort of extra add-on intellectual property enforcement mechanism of some kind or that it deals with issues in that space.
And so I think implementation really matters here, although the challenges and the risks that it tries to address are real.
Speaker 1
Okay, thank you very much. Next question is for Mr. Fiorillo. For the US economy and consumers, what are the potential ramifications of China's dominance in robotics?
Jason Fiorillo
Thank you, Mr. Chair, for that question. The impacts on the US consumer and the US economy related to China's dominance could be vast. Robotics is a critical technology for the US to build infrastructure, for the US government to function in public, private manufacturing sectors, and in governmental operations. If we do not have trusted robotics that we can allow to use in our facilities, in our commercial enterprises, in our government institutions, then China will essentially have a backdoor to robots that can autonomously walk throughout facilities, collect data, and possibly literally throw a monkey wrench into manufacturing operations. It is really important for the US economy that we actually provide this critical technology in a trusted way with freedom from backdoors and freedom from foreign involvement in this infrastructure.
Speaker 1
Thank you very much. Our next question is for Mr. Durbin. How does having a national framework in policy areas like data privacy and AI help maintain U.S. competitiveness?
Marty Durbin
Thank you, Mr. Chairman. The challenge that we see is that if we do not have a national framework, we have already seen in data privacy that more than 20 states have enacted their own laws. Now, there is some good news in that as many of those states are following the same consensus program, but not all of them. And I think that what we've seen is that without a clear national framework, the added costs that will frankly hit small businesses far harder, that have a harder time, that are, of course, engaged in interstate commerce, having a much harder time. In fact, we've been working with them, polling them to show that 65% of them show that they're very concerned about the regulatory uncertainty that comes about from state by state.
So the national framework will really help to address that issue.
Speaker 1
And we're working on it. So I appreciate that very much. I'm going to yield back now and recognize the ranking member Schakowsky for five minutes of questioning.
Janice Schakowsky
Dr. Eisler, I wanted to talk to you. And if you could talk about the ways in which the work that is being done right now is really not sufficient to make sure that we are going to, how can I say it?
Janice Schakowsky
Yeah, we want to see what kind of innovation is being done and enough is being done. And I know that you've been working on that. And I want you to talk about that. I want you to know that you have how many minutes? 13 total or something less?
Four minutes.
Janice Schakowsky
Four minutes. Okay. I'm going to be quiet, but talk about what you know is happening and how we have to do more.
Jedidiah Isler
Thank you, Ranking Member Schakowsky, for your question. If I could distill my testimony into one thought, it'd be that America cannot legislate technology leadership at the end of a pipeline without, while weakening the people and institutions at the beginning of it. So we've seen some of that happening already. A recent survey found that 22% of labs had rescinded offers to students, staff, and researchers, postdoctoral researchers. 53% of researchers had already advised their students to seek future education outside of the United States.
And the European Research Council saw a fourfold increase in applications from Americans for advanced grants. International graduate student enrollments have gone down in 2025. That's the largest single-year decline since the pandemic. And we're seeing that our competitors aren't waiting. China is the most dramatic example, but India, South Korea, and the EU are all accelerating investments in this academic market.
So if we're interested in figuring out how to ensure that we have a technology base that supports scientific leadership, then it depends on a chain of capabilities long before that product reaches to market. So when our federal research funding becomes unstable, or graduate students and postdocs lose opportunities, when international students no longer see the U.S. as a viable place to do train and build, and when federal agencies lose technical staff, the damage done doesn't stay confined just to the university setting. It shows up in fewer startups, slower commercialization, weaker supply chains, and thinner technical agencies in the government.
That's less ability to compete with other countries, not more, particularly those that are already organizing their entire systems around these technologies and the ensuing strategies. So while I think that the committee's work here is incredibly important, we can't treat any of these as standalone answers to questions. We ensure enormous benefits, some of which we've talked about here, world-class universities, national laboratories, extraordinary companies, and deep capital markets that people want to stay and build with. So my recommendation broadly across these bills is that we should ask the practical questions based on what we know about where we stand on investments. Does the bill strengthen research?
Does it build and retain talent? Does it give an agency the capacity to actually act, and does it help American discoveries become American products, companies, jobs, and strategic advantage? If the answer is yes, then the proposal is helping. We're helping to build the next generation of American leadership. If the answer is no, then we've got some work to do, and we need to name that any important technology without the underlying strengthening the underlying conditions will not allow us to lead.
So we want to ensure that technology translates the science and innovation into tools, products, and capabilities, and that we are translating the underlying ecosystem, the basic and use-inspired research, the technical talent, resilient supply chains, and strong federal agencies to keep up. So my message isn't simply to fund science, although I will not turn that down, it's that if you want Congress to lead on American technology leadership, we have to both protect and cultivate the ecosystem below it. So thank you for the opportunity.
Jedidiah Isler
I'd say no. We have sustained significant recession in terms of our talent capabilities, in terms of our federal science and technology researchers. We've lost over 10,000 in the last few years. That's PhD-trained science and technology experts in our own federal government who are no longer in those roles. That represents 14% of the total technical talent in the government.
So we are sort of going the wrong way in terms of talent.
Speaker 16
Our time is up, but thank you so much. I really appreciate it. I got the gist of that, that's for sure.
Great.
Brett Guthrie
The gentlelady yields back. Now I recognize the chairman of the full committee, Mr. Guthrie, for his five minutes of questioning.
Thank you. I appreciate it. The first question, Mr. Durbin, is for you, and I'll tell you your South Kentucky cousins say hello. The European Union's Corporate Sustainability Due Diligence Directive imposes significant burdens on American businesses, even if they lack a meaningful territorial connection to the EU. Can you discuss the impact of this requirement, what this requirement is requiring, and the impact on American consumers and workers?
Marty Durbin
Of course, well, thank you for the question, Mr. Chairman. Yes, what we're finding is that not only is this an increase in compliance costs, it is adding legal liabilities. And what's happening, even though the initial company that is affected by the rule might be a large company, unfortunately, it applies to their subsidiaries and their supply chains throughout the world. In many cases, that means that the trickle-down effect means that you've got smaller companies, imagine vendors or suppliers to a large company, or even startup companies that are trying to get into relationships with larger companies that simply don't have the wherewithal to either to comply with and to monitor all of the different parts of their supply chain. It has a real impact, especially on those smaller companies.
Again, it does affect the larger companies as well, and we do think that by having the extraterritorial impact here, it's simply the wrong way to go.
Speaker 14
We've been working closely with with the EU to try to to make appropriate changes.
Brett Guthrie
Thank you and one of the of us hosting the world this summer is all the Europeans who are most of them visiting our great cities and and they're obviously great so they they see that but now they're out in the towns and suburbs and are just amazed what you hear at the standard of living of America and so we need to make sure we understand what we do different than they do and not follow their path. So Mr. Chilson, the United States leads the world in developing the most cutting-edge proprietary AI models. At the same time we lag behind when it comes to open source AI. By almost every measure Chinese open source AI models developed by companies such as DeepSeek and Alibaba are being widely adopted around the world and even by American companies. So Mr. Chilson, why is it critical that the US lead in open source AI models and how can the Open Source AI Leadership Act address this gap?
Neil Chilson
So thank you for the question Chairman. Open source is critical for a bunch of different reasons to American competitiveness. The first is as a internally as a competition. You know the US economy and innovation is spurred by competition among providers of services and open AI open models and open weight models provide a key check on the centralization of model building within the largest companies and a great entry way for smaller developers, scientists and researchers to build new products and and develop their own open models using open source and open weight AI. And so the competitive issue is really important.
The the market as you pointed out is moving towards open source in many ways in open weights. We see this OpenRouter which is one of the sites that monitors how people are using AI has said that you know the use of open source tokens processed on there has has risen from 34% in January to 65% in June. Essentially as AI has become more popular more people are looking to open source models and on there the four most popular models are Chinese with DeepSeek in the top spot. And so that is a market that we cannot afford to cede not because of the competitive effect but also because of the cultural and the security.
Speaker 11
So what do we need to do to make sure we don't cede? And how will the act help?
Neil Chilson
Yeah the act that we're talking about today does a great job by giving Commerce an affirmative mission to look at this challenge to support qualified open models and to look at Chinese models and see where the security risks are because those models do raise risks when you use them and I think that having Commerce dig into these and do a report that talks about that is a good first step towards you know regaining open source leadership in AI.
Brett Guthrie
Alright but so Mr. Fiorello briefly what is the number one action in your opinion that Congress can take to maintain US global leadership in emerging technologies? Just one big thing you want us to do.
Jason Fiorillo
Yeah the number one thing that we can do is establish a US national robotics strategy. The bill before us today the National Commission on Robotics Act is a thoughtful, appropriate and very useful step in doing that in that it brings leaders of government, leaders of industry and leaders of technology together to establish this national framework for a very important technology that is used not only in manufacturing, in commerce but also in government as key dual-use technology.
Speaker 18
Okay, thank you. My time's expired and I will yield back. Thank you. Thank you.
Speaker 1
Now we'll recognize the gentlelady from Florida, my partner in getting COSA across the finish line on the House floor. Let's keep it going. So I recognize you for five minutes. Appreciate it.
Kathy Castor
Well, thank you, Mr. Chairman. Thanks to our witnesses again for being here today. This is a very important topic. In Congress, I've spent a little bit of time on the Select Committee on the Strategic Competition with China and then in energy on this committee and on the Select Committee on the Climate Crisis. And all of that taken together at this point in time, I truly fear that America is falling behind China, the EU, other economies when it comes to clean energy technology.
And that means that that is putting our workers, our communities, our businesses at risk if we allow this to continue. I liked your metaphor, Dr. Isler, about cultivation. And right now, if you think about clean energy technologies, our own government is stamping out the cultivation, all of that innovation that we want to own in the market, in the marketplace. The Trump administration has cancelled fully permitted clean energy projects and research on batteries, on solar, on wind. And then I know this week my Republican friends are going to celebrate the passage, the one year ago of the Big Ugly Bill.
Well, that Big Ugly Bill ripped away all of the investments in clean energy technologies and projects that was going to help counter the Chinese Communist Party and others across the globe, invest in our own business, invest in our own energy communities, manufacturing strategies for batteries, electric vehicles, geothermal power, industrial decarbonization, solar, wind, hydrogen, and all of the important supply chains that go along with it. This makes no sense at all to me. Especially at a time when everyone is focused on our competitiveness and we know that dangerous climate pollution is fueling higher costs in whether it's droughts or fires, heat waves, more extreme events and health impacts. Why would we cede our leadership role in the world on innovation and science and research in this area and allow other nations to dominate when the rest of the world is turning to these kind of clean energy technologies? So Dr. Isler, what impact does that have when we say through the Inflation Reduction Act, we're investing here, we believe we have to confront the climate crisis and we want to do it with the American worker and American innovation, and instead the rug is then ripped out from under our businesses and scientists?
That doesn't seem like a recipe for cultivation and dominating a competitive ecosystem, does it?
Jedidiah Isler
Thank you so much for those remarks. I appreciate your attention to detail and your service on these efforts. It is true that we are ceding significant leads in this competition with China. Just to give some real clear stats, particularly on the issues that you talked about, according to the Australian Strategic Policy Institute, between 2003 and 2007, the US led in 60 of 64 technologies and China led in three. Between 2019 and 2023, the US led in seven of 64 technologies and China led in 57 of the 64.
So we've seen an incredible shift in tide on topics across the domains that you spoke to. In terms of how this impacts researchers and how it impacts the innovation that comes from researchers, it's a disruption in the process.
If experiments are going on, if projects are permitted, if pilot programs or prototypes in the user facilities are disrupted, that in some cases can disrupt the results we want to see. It means that good products that are past prototype or even all the way out to commercialization, they're not able to serve the country in the ways that we've invested in them through this garden cultivation and full stack of R&D innovation. So we certainly recognize and want to continue to raise to the surface that the shift in our priorities as a nation have deep impacts on the results we can produce. And because science and technology and the transition of an idea from literally an idea in a researcher's head to something that someone can commercialize requires nearly uninterrupted investment and commitment from the nation all the way through.
Kathy Castor
And I meet so many young people who are inspired to do things like solve the climate crisis and develop some of these new clean tech answers. What signal does it send to them when the federal government here at the 250th celebration of our independence says, oh, you know what, we're going to cut that research and you don't have the same opportunity to find those answers?
Jay Obernolte
It sends a poor signal. I know I have only a few seconds, but it sends a poor signal and we could do better.
Thank you very much.
Jay Obernolte
Gentlelady yields back and I'll recognize now Mr. Obernolte for his five minutes of questioning.
Well thank you Mr. Chairman and this is a hearing I wish I had multiples of five minutes, but we'll try and go fast. Mr. Fiorello, we'll start with you. Obviously you and I share a belief that robotics and physical AI are going to play a key role in ensuring future American competitiveness and our economic and industrial future. In your testimony you mentioned my bill, H.R. 7334, the National Commission on Robotics Act, which would direct the Department of Commerce to establish an independent and temporary commission to examine domestic and international developments in robotics, supply chain risks, manufacturing competitiveness, and the role of robotics in economic growth and our national security. And I just, I know Chairman Guthrie asked you about this a few minutes ago briefly, I wanted to give you a little bit more time to explain why you think such a national commission is so important.
Jason Fiorillo
Thank you Congressman Obernolte. It's a real pleasure to be here and I appreciate the question. I think that it's absolutely critical for the U.S. to lead in this technology, which we refer to as advanced mobile robotics or physical manifestation of artificial intelligence, because it is going to be in every sector of our economy. Beyond that, it's going to be essential that we have a trusted supply chain of physical AI-enabled robotics in order to assist with our manufacturing, to assist with our logistics, warehousing, perimeter security, public safety, defense, transportation, and infrastructure. The robots that Boston Dynamics provide are used in a variety of ways to keep people safe and to keep public safe.
A commission like this will actually help us create a national U.S. robotics strategy, which will direct those that are not in government to work with people that are in government in order to create safety standards, technology standards, and a single point of connection to harmonize the technology leadership that we have in industry with the policy government we have here in the government.
Jay Obernolte
Well thank you, I completely agree and this is the first step in hopefully establishing a strategy for making sure that we maintain the lead in advanced robotics.
Mr. Chilson, appreciated your testimony. You highlighted the fact that just in the last six months we've had over 1700 bills introduced in state legislatures on the topic of AI regulation and I believe the direct quote was that we can't have one capital setting the standards for all 50 states. That's not really what our system of federalism is designed to do and in fact Article 1 of the US Constitution says that only one regulator, this body, the US Congress, should have authority over the regulation of interstate commerce. As you're aware, several weeks ago Congresswoman Trahan and I introduced what we call GAIA, the Great American AI Act, that would establish some federal standards for transparency and managing catastrophic risks in advanced AI models and pairing that with a limited form of preemption that makes it clear where the lanes for federal regulation are and where the lanes for state regulation are and what's preempted and what's not because it's going to require a partnership in our opinion between the federal government and the states to regulate this fast-moving technology. Do you agree with the approach that we're taking in GAIA?
Neil Chilson
Yes. The approach that you're taking in GAIA addresses acknowledges obviously that with a general-purpose technology there will be a role both for the states and for the federal government. But AI models in particular are inherently interstate. They are the types of investments that you're not going to be able to develop a model, a frontier model that is customized for California and one that's customized for Oklahoma. And so the approach that you take with a narrow preemption that focuses on AI models is I think makes a ton of sense both from policy but also constitutionally.
I think this is it is the clearest example of a technology that should be governed at the federal level and and and that where state state regulation will be counterproductive. Now there are lots of other applications of AI that are also probably interstate and where there should be guidance at the federal level but and and there are some that are clearly intrastate, things like the the state's own use of AI technology where federal overreach would be a real problem. But I think the approach that you take in GAIA towards preemption targeted towards AI models makes a ton of sense.
Jay Obernolte
Well thank you. This is just the start of this conversation. Let me point out that that preemption has is going to have policy oversight here in Energy and Commerce Committee, that title. So this is just the beginning of that discussion. But thank you for your testimony.
I yield back.
Speaker 1
I thank the gentleman. He yields back and now we'll recognize Representative Soto for his five minutes of questioning.
Darren Soto
Thank you Chairman. Science is foundational to America's success and our innovation is the envy of the world. Investing in research, strong peer review, all critical, then supporting advanced manufacture to capitalize on these ideas. But that's not enough nowadays. Artificial intelligence, advanced chips, quantum computing all are essential to boosting production and efficiency.
And why folks should care about it at home, one this means jobs, second it's how we help lower costs and of course maintain our world leadership. We saw with the CHIPS and Science Act it created a pathway of investment that was greatly helpful to Central Florida and places like NeoCity where we saw a half a billion dollars in investment over the next 10 years both through NSF, through the Army and through also the American Rescue Plan. And we've seen new players come into the the our tech facility including Elpis, which is a 470 million dollar new microchip manufacturing center. But to compete with China and Russia we need to continue to be investing and so NIH and NSF cuts that we've seen lately are are causing us to potentially fall behind. And then we see politicization in some of the grants with 7800 of them being cut lately which could result in a 1 trillion less in GDP.
But there have been some positives, like we see with the aerospace microchips we're making at NeoCity. I also remain concerned about advanced chips being greenlighted by the administration to be sold to China, including the NVIDIA H200 chips. Mr. Chilson, I know you're advocating for broad efforts on AI. Does the approval of these sales, is that something we should be concerned about, or is that something that you think is important to the evolution of AI?
Neil Chilson
So competition with China is not only about chips. It is about the complete stack of components. And additionally, we have to be wary of creating a stronger incentive, seeding the entire market to Chinese purely industrial, their own capacity, where they will build without competition from us and then subsidize those products as they sell them around the world. And so when we look at a company like Huawei, who is building very capable chips right now, they're not nearly as efficient as our leading chips in the US, but they are capable, and they will be subsidized by that government. We don't want to seed that market, not just in China, but internationally.
And so I barriers to selling chips into China is there's a balance to be struck there about both of those concerns: cutting them off, but also over-incentivizing them to supply their own chips.
Darren Soto
Thank you so much. Mr. Furio, I appreciate the great work that you all are doing in Massachusetts and other areas on US robotics. And a lot of us saw this viral robotics video a while ago from China where they were doing martial arts. You've mentioned in your testimony about military parades. All that's very fancy, but I'm most worried about manufacturing.
How do we measure up right now to China with regard to robotics manufacturing?
Jason Fiorillo
So thank you, thank you for the question. I'd say that the answer is complicated. When we talk about advanced mobile robotics, what we have is great technology, amazing AI, fantastic sensors, but the US lacks scale. So what we need right now, we're sort of in a chicken-and-egg scenario where we need demand in order to start cranking the flywheel so that we can actually start manufacturing at scale. And we need domestic and friend-shored supply chains so that we can actually produce in volumes that allow costs per unit to go down.
If we do those two things, then I think we can beat China any day that we want.
Darren Soto
Now, would the demand be incentivizing major companies to utilize this advanced robotics for manufacturing? So you think it's not moving along as quickly as it could?
Jason Fiorillo
Absolutely. I think it needs to move more quickly in order to bring more robots into more facilities to do more things.
Darren Soto
And then you said advanced mobile robots are the physical embodiment of AI. What a fascinating proposition. Can you give us one or two examples you think you'll see of the embodiment of AI in robots over the next five to ten years?
Jason Fiorillo
Oh, absolutely. So we make a quadruped robot, and we make a humanoid robot, and we make a box-moving robot. Those are three different types of robots that all have three different types of AI models that allow them to touch, see, and interact with the world. Beyond that, you could actually use the AI that's on board those robots in order to poll them and ask them questions about what they have seen and then take action when they see something else that's interesting, unique, startling, or important. So the AI models that we are developing are based upon the robot's experience in the world, experience in the manufacturing sector, and perhaps even videos of other robots doing the same things in order to sort of leverage these learning algorithms and make the robots themselves much more valuable.
Incredible. I yield back.
Speaker 1
Thank you. Gentleman yields back. I now recognize Mr. Bence for his five minutes of questioning.
Speaker 19
Thank you, Mr. Chair, and thanks to the panel for being here. So you all, all of us saw the outcome of the elections of the last couple of weeks, and you may have noticed an absolute illustration, clearest possible illustration of fear of artificial intelligence as different candidates ran against data centers. And I was looking through your materials and I was looking fruitlessly for remarks regarding what we should be doing about this fear. So I'm going to ask each of you, I'll start with you on the end, that this Congress should be doing something to breathe confidence back into the American people that somehow we know how to control AI, because they don't believe it. And I think they have a mental image of each of these black boxes containing some Frankenstein-like doctors putting together some fearsome monster they can't possibly control.
The analogy I heard of we here in Congress was people on donkeys trying to install a seatbelt on a Tesla going by at 100 miles an hour. That's not illustrative, is it, of confidence? So what I would like you to do is share with us how we would convince the American people we know what we're doing when it comes to the very devices that the materials and the bills today suggest that we facilitate the creation of. Starting with you.
Jason Fiorillo
So thank you, Congressman, for the question. So I admit that actually any new technology can be frightening for people that don't understand it, and we certainly owe a duty to the American people to explain the technology, demonstrate how it is helpful, and demonstrate how it can be used. Certainly in robotics, there are plenty of examples of frightening or scary robots, but there are also plenty of great examples.
Speaker 19
Well, how would you do this? How would you do this? What's the, what tools would you use to convince the American people of what you just said?
Jason Fiorillo
So I think that they need to see the robots. And so we do a lot with Boston Dynamics. We try to expand online and show people.
Speaker 19
Don't you, don't you think that robots are going to be looked at as job-stealing devices, that the more robots we have, the fewer jobs? What I'm getting at here is this is a huge problem, and I haven't seen the proper focus either by perhaps we here in Congress, for sure we here in Congress, and those of you in technology. Mr. Hilson, your turn.
Neil Chilson
Sure. So the Abundance Institute, we, we not only care about policy, but we care about culture as well. And the culture of America right now is one that fears our cutting-edge technologies, in a way that is actually an outlier compared to the rest of the world. So the Chinese believe, 80% of Chinese believe AI will make their lives better. Right now in America, only one in five Americans have actually used a chatbot to do a significant task.
Lots of people have tried them out, but not really used them in their day-to-day practice. So I think getting familiar with the technology is key. I think also the tone by which Congress talks about it is also really important.
Speaker 25
But your solution, it's nice to say that, but tell me how would you make that happen?
Neil Chilson
So, so there's lots of ways we could do that. I think the way that Congress talks about this as an opportunity is one thing that can happen. We are engaged at the state level, interacting with companies to try to demo.
Jedidiah Isler
Thank you for that question. I think zooming in on what we could do, first I think we could build out and and reinforce community benefit agreements where we're talking directly to constituents about what their concerns are to be able to address them.
Speaker 19
You mean perhaps having these data centers do something locally that would promote some local thing and thus breathe confidence, create confidence in local people.
Jedidiah Isler
That's right. It it appears that the data center conversation is an intensely local one, a local jurisdictional one. That's how people engage with it.
Marty Durbin
I think it's broader than that, Mr. Durbin. Now just to build on that, I think I think you do have to address this at the local level. There are legitimate local concerns that have been raised, but there are good answers to them as well in in how industry is coming in to be able to say here's how we're going to address the challenges you see, whether that's on you know on jobs, on energy, on water, on how you know what type of Excuse me if I may, I've been going through the materials here fruitlessly for that were prepared for today's hearing, some indication, some recognition of the concern and fear that people have. I haven't found it yet, it may be in here somewhere hidden, but I think it's one of the biggest problems we're facing today back in Oregon.
We have people saying no, we do not want data centers. And yet at the same time they're busy using some of them, chatbots on their phones. There seems to be a total disconnect between the everyday application and and the need for these facilities. Mr. Chair, I think it's a huge problem, but I welcome this opportunity to discuss it. Yield back.
Speaker 1
Thank you, gentleman yields back. Now recognize the ranking member Mr. Pallone for his five minutes of questioning.
Frank Pallone
Thank you, Mr. Chairman. I have three questions for Dr. Isler, so I'm going to try to issue the answer them so we can get to all three, all right? The last 250 years of American innovation has seen incredible growth and prosperity, and many of America's greatest achievements have grown from public investments made in research and talent, especially over the last 100 years. Unfortunately, the Trump administration is turning off this engine of American leadership and global competitiveness. The administration has slashed billions of dollars in critical research funding, damaged the development of America's future scientific talent, and tried to shut down the institutions training America's advanced manufacturing workforce.
So my first question, Dr. Isler, is what is the impact of funding cuts already imposed on America's research institutions, and what are the consequences for our ability to compete with countries like China that are aggressively expanding their research investments? Maybe a minute or so.
Jedidiah Isler
I'm going to turn on my mic first. The short answer is we've already seen a reduction in the number of international students that are coming to the United States, we're seeing an increase in number of American grant applications to our allied nations, and we're seeing a reduction in the federal S and T workforce inside the government. So all of it we're seeing, all of it means we will not be prepared to continue to drive innovation.
Frank Pallone
Thank you. And let me get to the second one. On May 29th, the Trump administration proposed the most significant overhaul of federal grant rules since 2013 with direct implications for NIH-funded research. And that was tying federal grants to political criteria, which I think threatens the merit-based independence that made America a global science leader. When I was growing up, everything was merit.
Now you think it's, you know, all politics. So Dr. Isler, NIH funding priorities and review criteria are typically transparent. What happens to American innovation if funding decisions are made by non-scientific political appointees based on unclear criteria and shifting political priorities?
Jedidiah Isler
It's a good question. So first, you're right, the data bears out the exact thing that you remember. The American public has for a long time trusted the credibility of scientists, technicians, technologists, above basically any other field for a very long time. It's only in recent years that we've seen that trust decline. So if our federal agencies, particularly those at NIH as you point out, begin to make less transparent, very consequential decisions, it will erode our ability to provide American people the information they need.
We still see people trust in what science and technology produces. Seven in ten people across the country say that science and new technology benefits them. And so if they don't see their investments and they don't see the decision tree for what they're doing, then that confidence erodes and then the expertise is less useful. I did want to take the opportunity to your point about changes to the proposed OMB rule 2 CFR 200 to talk about the broad scope of...
I do have a third question.
Speaker 30
You ask your third question, I'll sneak my other one in and then you can get to that.
Frank Pallone
So this hearing features bills that address important sectors and technologies for American economic leadership. Democrats have fought for the future of American innovation in technologies like those with historic investments, including the CHIPS and Science Act, to essentially reinvigorate and modernize critical domestic sectors. But that progress is not self-sustaining. So my question, Dr. Isler, what additional steps must Congress take to ensure America does not cede its leadership in the technologies that will define coming decades, if you will?
Jedidiah Isler
I'd say the easiest way to think about that in the context of the bills under discussion today is to ensure that the studies, the commissions, all of those procurement restrictions are all tied to implementation pathways. So once we do the study, how does that support the workforce? How does that ensure innovative procurement rules? How do we make up the ground that we're trying to protect when we are slowing or in other ways blocking our engagement with other nations? So I'd say the thing we have to do is tie the idea to some implementation pathway that supports the full ecosystem: the researchers, the federal scientists, the user facilities, the prototyping, early capital investment, and then the actual implementation.
Frank Pallone
Well, thank you. And you know, I just wanted to comment on Mr. Chilson, this isn't a criticism, but you know, you talk about China, people like innovation, people, you know, like AI, whatever. The difference here is that we are a capitalist society, right? You don't, you don't get a job, you don't get a check, right? If you lose your job, you lose your house, you lose your health insurance, you lose everything, right, you have kids, right?
I mean, the problem is that if you're in a society, and I'm not saying I want to be in a communist society, but if let's use a socialist society, Japan, France, United Kingdom, right, most of those things are guaranteed. And so it's a little different. You know, when somebody thinks they might lose their job here, they lose everything and they don't have a livelihood. If you lose your job in some of these other places that are more socialist or even communist, you know, you might say, well, okay, I'll find another job, you know, the state will assign me something, they'll give me a housing allowance, they'll give me food. So Americans naturally are suspicious of change because they don't have any, they don't have the stability that exists in these other places.
So it's a little different in my opinion. And I don't think you can ever convince Americans that these changes are good given that their livelihood depends on their job. Thank you, Mr. Chair.
Speaker 29
Gentleman yields back. Now recognize Mr. Evans for five minutes of questioning.
Evans
Thank you so much, Chairman, and of course to the ranking member for this important hearing. And thank you to the witnesses for coming and testifying as we talk about everything that we need to do to make sure that America continues holding the leading edge in critical sectors from semiconductors to AI to quantum. In your testimony, you say that the open source AI gap is one of the most important issues in AI policy. Mr. Chilson, first question will be to you.
A discussion draft of my bill, the Open Source AI Leadership Act, would direct the Department of Commerce to identify and address barriers to the adoption of American open source AI models while also reporting on the risks of open source AI models from places like China. And unfortunately, China is dominating in this open source AI race, and we know that as they do in everything, they will leverage their position to rob American intellectual property, replicate that technology at a lower cost, and replace America's share in the global market through heavily subsidized products. And they also have laws on the books to require mandatory backdoors for the PRC and the PLA to be able to manipulate this technology. And so, two-part question to you: how did the Chinese get ahead, and can you give me a specific example of what happens if they stay ahead in the open source AI space?
Neil Chilson
Sure. So, the Chinese got ahead in the open source space in part because they had no alternative. They cannot keep up with the frontier models that we build. They have not been able to figure out how to do that yet, and we make the most amazing closed models, but they're within the services of the companies who are building them. And so they focused on open source.
In some cases, they leveraged, you know, the access to some of our models to be able to build some of those models. And so I don't want to diminish their innovative, incremental innovation in some of these spaces, but a lot of it was building on American innovation in the first place. And so that is how it's their focus on this technology that has gotten there. And a product that can be distributed for free is quite popular, it will turn out. So, so that's the other way that they've gotten ahead there.
So, and I forgot the second part of your question.
Give me a specific example of what happens if they stay ahead.
Neil Chilson
So, one of the biggest challenges is that every startup across the world, including American startups, will, when they are working at the garage level stage, when they're still trying to prototype, they will reach out there for the best free model that they can get, low-cost model that they would get. And if that is always a Chinese model, they will start to, they will start to gain market share upstream. As China does build the capability to develop these models, they will continue to creep into the market share, I think, further up the stack. And so that would be my biggest concern.
Evans
And that's a great transition to the next question. Mr. Durbin, can you talk a little bit about that? What is going to happen if China is leading in this open source AI race? What's going to happen just to our economy if they are getting in early for some of our growing and starting companies? And can you just talk a little bit more about the impact that's going to have to American companies if they're having to utilize open source Chinese AI?
Marty Durbin
Well, thank you, Mr. Evans, for the question. And I'll delve into this a bit. I'm happy to follow up with you in writing afterwards as well. But I did want to say that one of the things we need to focus on is the fact that we know that technology leadership is what will end up attracting investment here. And so all of the bills that we're talking about today, we know that that has that as a kind of a through line.
That's where we're focused, so that we're attracting the investment, maintaining the leadership, making sure we have the workforce driving toward the economic growth that this country needs. But happy to follow up with you separately on your question.
Evans
Thank you. And then in my final 50 seconds here, last question is to you. Today we're also talking about the Memory Chip Competitiveness Assessment Act, excuse me, by my colleague, Rep. Miller-Meeks. And I've heard from companies across the economy about the rising prices of memory chips. DDR4 memory prices have reportedly increased by 7 to 800 percent year over year.
What are the downstream impacts of these price increases on other industries and ultimately to the American consumer?
Marty Durbin
Well, there's no question that when we see an imbalance here, supply demand has been a greatly increasing demand for these chips across the economy, from yes, data centers, but cars, every other connected device that we have here. So the challenge now is to make sure we can keep get supply back, you know, back in in balance here. So whether or not there there may be opportunities to look at, you know, being able to ensure the availability of of chips, I think what we really need to do is is to make sure that we've got policies in place that will allow for the increased supply, everything from permitting reform to to regulatory policy to, you know, just ensuring that we've got the workforce needed.
And that brings down the price of anything with a chip in it. Yield back, Mr. Chairman.
Speaker 28
Gentleman yields back, now recognize Mr. Mullin for his five minutes of questioning.
Kevin Mullin
Thank you, Mr. Chair. Thank you all for being here today. My district in California is a shining example of U.S. scientific leadership. Known as the birthplace of biotechnology, the region and specifically my hometown of South San Francisco is home to a concentration of world-class research institutions, small and large biotech companies, and investor groups. Every day these companies are developing the next life-saving cure.
Without U.S. investment and research funding through agencies like NSF and the NIH, the U.S. biotech industry would not be what it is today. I am deeply concerned to see the Trump administration's continued attacks on research funding. From attempting to block NIH funding and more recently by removing all members of the NSF's board of advisors, this administration is anti-science. At the same time, the White House is not doing enough to address the ongoing memory chip shortage that could lead to significant price increases for consumers and negatively impact critical industries from broadband providers to medical device manufacturers. I am, however, encouraged to see ongoing bipartisan work to ensure we have effective regulatory oversight and ensure greater industry collaboration in the biotech space.
Specifically, we must ensure that bad actors cannot get their hands on dangerous viruses that have the capacity to cause widespread harm. Gene synthesis companies like Twist Bioscience in my district are leading the way in industry-led guidelines to screen orders and customers. However, studies have highlighted the concerning limitations of companies' screening processes as robust as they may be individually. Government participation is needed to ensure we are best protecting the health and safety of Americans while still allowing this industry to grow. So, my question, Dr. Isler, you mentioned in your testimony about the importance of ensuring that competitiveness and safety are designed together when considering biomanufacturing and biosecurity.
Can you elaborate on why government regulation in collaboration with the scientific community is not only necessary to protect public health and safety, but also helps preserve American leadership in that space?
Jedidiah Isler
Thank you so much for the question. It's a very good point. We see the technology is moving quickly and in order to make sure we can translate these innovations into a competitiveness, company companies need clarity, the public needs trust, and the government needs enough technical understanding to distinguish risk from speculative true risk from speculative risks. In that sense, you need scientists to understand what your core problems are, where the areas of concern are, and how to move forward well. The analogy we like to think of is that if you have a fast car, you need good brakes.
You don't build the car for the brakes, but the brakes help you have confidence to move fast. We think that smart regulation on some of these emerging technologies and biosecurity in combination with the researchers that are actively pursuing some of these frontier techniques are the way that we can tie protection and promotion at the same time.
Kevin Mullin
Thank you for that. And I just wanted to touch on proposals like the Biosecurity Modernization and Innovation Act are important steps in developing gene synthesis regulations that work for companies and protect everyday Americans. I want to shout out Mr. Pfluger for his leadership on the other side of the aisle on this bill. And while I have a few concerns about the text, including ensuring agencies like HHS are working in tandem to develop screening requirements, I look forward to working with everyone on the committee there after this hearing to strengthen the draft of that bill. So through bipartisan commitment to protecting research funding and developing common sense regulations that balance innovation and safety, the U.S. can continue to be a global scientific leader.
And with that, Mr. Chair, I yield back.
Speaker 1
Gentleman yields back. Now I recognize the vice chairman of the subcommittee, Mr. Fulcher, for his five minutes of questioning.
Russ Fulcher
Thank you, Mr. Chairman. Mr. Chilson, I'm going to deviate a little bit because the questions I was going to ask have kind of been addressed, but they brought up other ones in my mind. You know, historically in the tech sector, we've been able to at least rely to some extent on patents, on copyrights, arguably not as much as we'd like because they have a tendency to get stolen. Intellectual property has a tendency to get stolen, mostly by our adversaries. But that leads me to the question regarding AI.
To what extent is that even applicable in this new world? I mean, can you patent, can you copyright, and or is that just defeating the purpose of the technology in the first place with AI? Can you address innovations with AI and if and how that applies to be patentable, copyrightable, etc., and protected that way?
Neil Chilson
Well, there are many intellectual property issues surrounding AI. If I take your question properly, you're talking about how that could provide incentives to produce and maybe add to American competitiveness.
Neil Chilson
So there I am not a patent attorney, and there is a whole bunch of complicated issues around software patents. But the ultimate goal of our copyright and patent systems is to promote the useful arts and sciences. That's the phrase in the Constitution. And so that has to be the metric by which we measure. Right now, we do not have an innovation problem in AI.
The U.S. is driving ahead on that innovation side that I talked about, the three capabilities. That innovation component, we are the leading country in the world on building the best AI. And so the incentives that we might add there might be around applications, it might be around other types of issues, but I'm not sure that that is the lever that we necessarily need to pull. We might need to look at how intellectual property is affecting the ability of companies to build AI, but it's not necessarily the lever that needs to be pulled, I think, to drive innovation in this space where the U.S. is doing quite well.
Speaker 33
We've got a chief legal officer from a robotics company. What's your take on that?
Jason Fiorillo
So thank you for the question. I agree, it's a complicated issue. I do see that patents as a sort of living protection, and it changes over time. Anything under the sun that is made by man used to be the standard for something that could be protected with a patent.
Beyond that, it must be new, useful, non-obvious. And so patents that could cover AI techniques, AI applications, AI uses, or combination of AI and robotics, for example, ought to fall within statutory subject matter for the patent office. With respect to copyright, again, we have the human requirement. There needs to be a human involved to create a creative expression or something that can be reduced to a picture or a song or a speech like I'm making today.
Jason Fiorillo
Enforcement will continue to be a struggle, but actually the rapid rate of change of AI is also complicating and confounding this issue. So the technology moves so quickly that the patents often can't keep up.
Speaker 35
Got it. Okay, thank you. I'm going to shift gears a little bit. Mr. Durbin, I'm going to steer this to you just because you might be the most appropriate person to send this to. But you know, we have a dependence on supply chain with a lot of our adversaries.
I'm thinking the semiconductor industry because I've got a little bit of personal experience there. But we're very dependent on Netherlands, Germany, Japan for a lot of that infrastructure equipment for semiconductors. To what extent, if any, do you think we should be involved with preventing those allies from supplying our adversaries? It's one thing for us to have our domestic suppliers only purchase or sell to friendly nations. What about on the other side of that?
Should we be trying to influence our friends from supplying our enemies?
Marty Durbin
Thank you for the question. That's a good question. I'd actually like to follow up with you separately on that to dig in a little deeper. I do think those trade relationships we have with those allies are critically important for the technologies that we're developing. We don't have all of the inputs here.
We need to have those strong supply chain relationships with again, certainly with our allies. I get your question.
Russ Fulcher
I was unfair in asking that because I didn't have enough time. So I will follow up with you later. That was too big of a question for a few seconds. Sorry, thank you.
Speaker 1
Gentleman yields back. Now recognize Ms. Kelly for her five minutes of questioning.
Kelly Armstrong
Thank you to the chair and ranking member for holding this hearing today, and thank you to all the witnesses for taking the time to be here. Under the Biden administration, Democrats made historic investments in scientific research and high-tech manufacturing. The CHIPS and Science Act made strides in advancing the United States leadership in technology, and I'm very proud that my district has been a beneficiary. Because of investments and incentives of the CHIPS and Science Act, PsiQuantum is building a quantum computer center on the old US Steel site on Chicago's South Side, which had been vacant and unused for years and years and years. Now my district in Illinois are poised to lead in quantum computing and bring world-class innovation and jobs to Chicago's South Side.
This is just one benefit of the investments made by Democrats. Unfortunately, this administration has shifted course. Now the US is lagging behind others. President Trump has gutted programs that exist to keep America competitive and a leader on the global stage. In 2025 alone, this administration cut or withheld over three billion in NIH and NSF grants that would have funded important scientific research.
The administration has also proposed shutting down the Manufacturing Extension Partnership, which supports small and mid-sized manufacturers, by issuing notices of non-renewal to nearly a dozen state programs before congressional opposition led to the notices being rescinded.
We've heard a lot from all of you about the importance of leading, innovating, and remaining globally competitive. However, the United States cannot realistically expect to outcompete China if we keep cutting programs that support investments in scientific research, workforce development, or manufacturing. Dr. Isler, what message does the administration's repeated attempts to roll back this funding send to researchers and to companies who are deciding whether to invest in the United States?
Jedidiah Isler
Thank you so much, Representative Kelly, for your question and your comments, and kudos to the work happening in your district. You are correct, right? We've already seen the brain drain. We're seeing reversals in projects, in people, and in resources to continue to turn these innovative ideas into actionable business and capital investment. We're also seeing the significant reductions in federal science agencies are cutting off our ability to do that middle-stage prototyping to allow things to move from that prototype to the market.
So at basically every part of the R&D infrastructural stack, we're seeing challenges. These challenges increase the cost of doing business, they increase the risk of investing in these early-stage technologies, and they decrease the likelihood of success. So if we're unable to provide a strong baseline of support across the ecosystem, then all that we invested all along that stack then sort of ends up in a place where there's no one that is able or interested in investing when we need it most.
Kelly Armstrong
And what lessons from the CHIPS and Science Act should Congress apply as we take future steps to support additional investment in advanced manufacturing, research, and workforce development?
Jedidiah Isler
One of the things we found so impressive about the CHIPS and Science Act was the hand-in-hand approach to the near-term problem, which was how are we going to bolster our manufacturing infrastructure? And we've already seen gains on that from that legislation to now, even with some of the challenges you face. But on the other hand, the "and Science" part talked about reinforcing the R&D infrastructure, to your point, workforce incentives, capital incentives, and the like. And so in that bill, having both reinforced at the same time made it so that you could have this virtuous cycle, you could cultivate your garden, to continue that analogy, in a way that really supported innovation so that when you got to the end, there was a ripe and robust commercialization avenue. And so I'd say that's one of the things we learned.
I think one of the harder lessons in retrospect is that if we don't sustain that investment, even the gains we made just a few years ago can evaporate almost immediately. And as I mentioned earlier, that getting back on track takes much more time than the retrenchment. So there are many things to learn, but I think as long as we're combining the near-term solution with the ecosystem and a pathway for implementation, we're setting ourselves up for success.
Speaker 38
Thank you so much, and thank you to all the witnesses, and I yield back.
Speaker 39
Thank you very much. The gentlelady yields back the balance of her time, and the chair now recognizes the gentlelady from Washington for five minutes for questions.
Shrier
Thank you, Mr. Chairman. Thank you to our witnesses. The number one issue that I'm hearing about in my district from my constituents from all over the district is affordability. And working families are seeing their costs go up across the board, from gas to groceries, you all know this, wages aren't keeping up. One of these factors that's raising prices indirectly is actually the rapid growth of AI.
You might think the opposite, that's where all of we would all be inclined to think everything gets more efficient, it gets less expensive, but it turns out that the ever-growing demand for memory chip capacity for AI data centers is far outstripping supply. And this is directly increasing the cost of memory chips in other industries, particularly for consumer tech like phones and laptops and game consoles.
And at the same time, all of those devices are going to need more and more RAM in order to allow their users to optimize the use of AI on those devices. And so there's this escalating demand and then a supply problem. Just last week, Apple announced a significant increase in prices for its most popular products due to the high cost of memory chips, and this is from a company that has generally prided itself for price stability and its own supply chains. The cost of memory storage is going up whether there's a data center in your backyard or not, and this shortage is going to impact consumers all over the country and the world. So that's why I'm glad to see this committee consider legislation on memory chip competitiveness, and alleviating this shortage is critical to provide affordability for Americans and to protect our competitive and innovative edge.
Dr. Isler, thank you for all of your comments today. In your testimony, you referenced the Memory Chip Competitiveness Assessment Act and the need for an assessment of U.S. exposure in the memory chip marketplace. I've seen some experts say that the sky-high cost of memory storage may not start to come down until 2028. What should the Department of Commerce be exploring to provide some short-term relief in the market?
Jedidiah Isler
Thank you so much for the question. There are benefits to a study and benefits to looking into the question, and we think some common-sense ways that we can produce specific legislation vehicles that support keeping that price down could include some trusted procurement requirements for sensitive federal applications to ensure that there's a market for products, a demand backstop that allows us to put a floor on the downside impacts on the market so that domestic memory investment is incentivized instead of discouraged. We think scaling up loan guarantees or patient capital in ways to sort of acknowledge that it's going to take the years that you talked about, and the third, fourth is making sure that we are investing R&D next-generation chips, excuse me, investing next-generation chips investment in memory as well.
Shrier
I appreciate that. And all of those things are so similar to what we talk about in other areas of need, that if you just make sure that you provide the marketplace and some assurances for production, the manufacturing will come. I'm wondering while we're on this topic, I have about a minute left, are there other critical components as you're looking into the future where we may also need to quickly focus on shoring up our supply chain or domestic manufacturing to avoid a similar problem?
Jedidiah Isler
Yeah, one of the things that keeps us up at night as an organization is an independent supply chain, particularly as it relates to critical minerals. So there are a number of critical minerals that we don't have the geologic benefit of having, but there are researchers that are looking into what the next generation of a semiconductor materials and minerals could be. So I think looking around the corner, investing in science, the basic science of like what other minerals could produce similar results would be a good one.
Shrier
That keeps all of us up at night too, and thank goodness for our national labs who are doing exactly that research, and I will say that is yet another reason our country needs to quintuple down on scientific research. Last just really quick question, Mr. Durbin, I'm curious to know what you're hearing from your member companies on the memory chip demand, and are you, what are you hearing from companies about like struggling to source these chips?
Marty Durbin
Well, again, as noted earlier, I think there's no question we've got a, the demand has just continuing to increase and the challenge we have is how do we now meet that demand and increase supply? I think by doing that we can again everything from encouraging permitting reform to specific investments to incentivizing innovation. I think this is another area where we're likely to see improvements in innovation of the chips themselves that will help to bring down costs.
Thank you. Yield back.
Speaker 46
Thank you very much. The gentlelady's time has expired and yields back. The chair now recognizes the gentleman from Texas's 12th district, five minutes for questions.
Speaker 42
Thank you, Mr. Chairman, very much. Thank you to the entire panel for being here today. Mr. Chilson, your testimony notes the importance of distinguishing structural shortages from normal market cycles. Do you believe today's memory shortage represents a structural shift driven by AI demand rather than a temporary supply disruption?
Neil Chilson
So in short, yes, but the memory market is complicated. It's not just one market. We have, we have to distinguish between high bandwidth memory, DRAM that's tied to AI, you know, solid state drives. The consumer side issues that Rep. Shrier was talking about are not necessarily the same types of memory, but they are affected by the investment that companies are putting into or the opportunity that the memory producers are seeing. And so AI has changed and I think because the AI demand is great and growing, I think we will continue to see a change in what the memory market looks like.
In order to meet that, it's not going to be enough to sit back and wait. The companies are investing, the manufacturers who build these products are investing, but it's a long process, especially with permitting and in particular, I would say energy is the probably the longer term constraint. Short term, we're looking at, you know, the packaging facilities, building those out and those run into all the challenges of building a new physical plant. But longer term on the 10-year scale, it's energy that's going to be a constraint here.
Speaker 42
Thank you. And Mr. Isler, the BEDROCK Act and Biosecurity Modernization Act both direct the Commerce Department to assess supply chain vulnerabilities and commercialization barriers in biomanufacturing. From a scientific standpoint, how serious are those vulnerabilities today and what role should the federal government play in accelerating the domestic commercialization of biomanufactured products without stifling private innovation?
Jedidiah Isler
Thank you for the question. It's a good one. I think picking up on some of the threads that I mentioned earlier, it is our responsibility in a rapidly changing biosecurity and biotechnology landscape to both promote innovation and protect American citizens as we do so. The BMIA and BEDROCK are really good examples of the balance between the two. They're two sides of the same coin.
Often when we think about biosecurity, we think about health-related or public health effects, issues like the Ebola outbreak in Africa or the screwworm virus wreaking havoc in Texas cattle right now. They're two good examples. But that's one portion of the biosecurity landscape that we need to address. There are consequential examples of biosecurity concerns in energy. It fuels biology now, excuse me, biology now produces new fuels.
There it is. And we also even store data with biology. So biosecurity extends outside of public health concerns and into other domains for sure.
Having essential and critical responsible regulation allows us to actually answer the problem faster. So now if we have a problem on our cell phone, for example, we get an upgrade overnight. It's because of cybersecurity protocols. That's what allows us to do that. Right now in biosecurity, particularly if you're talking about outside of Health and Human Services, you do, outside of medical constraints, we don't have that same infrastructure to support that.
So I'd say that on the health, the medical, and the public safety perspective, we need HHS, we need NIH to continue their leadership. But we also need agencies like Commerce and NIST included, USDA, DOE, DHS, and OSTP as security agencies to be weighing in on these other domains of biosecurity. So having an integrated approach where we're bringing the best of all these agencies into conversation to ensure that we can both innovate, we can identify threats, and then we can address them quickly without getting into territorial debates is the way to move forward. So that's why we're trying to consolidate and streamline biosecurity.
Speaker 42
Thank you, Doctor. Mr. Chilsom, in considering legislation to secure U.S. technology leadership, what is your opinion the greatest federal bottleneck that most threatens U.S. AI leadership right now? Power generation, permitting, grid interconnection, or regulatory fragmentation, and what should Congress do to change or remove it?
Neil Chilson
So, thank you for the question. I think the biggest concern right now is, I talked about the three capabilities that we need in technology leadership. Deployment is the number one challenge that we face. Getting technology out there to the many victory gardens. I love the garden metaphor as well, but getting it out to the many victory gardens of the entrepreneurs who are going to use these technologies and tools to produce real value.
That's our biggest challenge, and I think there's a lot Congress can do to remove the barriers that are stopping tech from getting out to people.
Thank you, sir. Thank you, panel, very much. I yield back.
Speaker 43
The gentleman yields back, and the chair now recognizes the gentleman from Texas' 33rd District for five minutes of questions.
Marc Veasey
Mr. Chairman, thank you very much. I think that all of you on the panel know that AI holds a lot of tremendous opportunities, but it's also creating a lot of anxiety, particularly amongst college graduates. I know that I have a son that's in college that's very worried about AI right now. I met a woman that works here in the D.C. area whose daughter graduated from SMU recently with a CompSci degree. And now, when the kid went into, graduated high school four years ago, it was like, CompSci, the world is going to be yours, and now it's like they can't find a job.
And people are concerned about that. And I wanted to ask Mr. Chilsom and Mr. Durbin, and just a very brief answer if you could. Do AI companies have more of an obligation to their shareholders or to the general welfare of the U.S. public?
Neil Chilson
So as a computer scientist myself, I've seen the power of these tools, and I think there's still a lot of excitement. I would encourage people to still go into computer science. I still think it's a very valuable field. Companies have an obligation to create products that solve their consumers' problems. And that's their primary goal, and I think that America has been very good at that.
Speaker 45
But as powerful as this tool is, and as disruptive as it can be to the job market and the U.S. economy, when you have a tool that's that powerful and can do other things right outside of the economy, we're already hearing all these things about what Claude can do. Do these companies have a greater obligation to U.S. safety and just general welfare than some other typical American good or product does?
Neil Chilson
So I do think that the general purpose nature of the technology does mean that it will have a giant effect. It means the companies need to be very conscious of what that effect is.
Speaker 41
You think but with that effect do you think they have a greater obligation to their shareholders or to the general public?
Neil Chilson
I think it means that their market is enormous and their customers are enormous and they have a obligation to their customers.
Yeah.
Marty Durbin
I'd argue that it's a shared obligation between the between those two, but I also think that the companies that are involved in this are you know not only acknowledging leaning into the fact that they understand there's disruptive parts of of of the technology and what this is going to mean for for the workforce. Our Chamber Foundation just a month and a half ago had an you know day-long day-long session on on workforce implications of AI and essentially the the baseline question was as as you posed it, what's going to happen to my children? So I think they're you know leaning into how and by the way I think there are other there's studies coming out I just heard a story this morning out of Princeton where they're starting to find that you know the implications aren't as dire as as some are fearing. I'm not saying that everything's fine. I'm just saying I think that we're starting to see that we may find ways that the the technology will not just destroy jobs, it will actually create new opportunities for other jobs.
Marc Veasey
Yeah, just like with these data centers. I mean the way that people think about mergers and the way people think about antitrust could certainly change greatly if they feel that their prospects for their kids and themselves being able to have some upward mobility changes greatly. Dr. Isler, as memory chip supply tightens, the concern is not just whether these companies can build the next generation of AI products, but whether families can afford the everyday technology they rely on. How can Congress make sure that continued access to affordable consumer devices such as computers and routers and connected home devices and medical equipment as demand for AI chips increases? We'll love to know that from Dr. Isler.
Jedidiah Isler
Thank you so much for the question. It's going to be important that we identify these critical chip shortages because we know that they will drive prices up and out of the the reach of most of our society. We think production requirements tied to federal funding are is a good place to start. We think trusted procurement that sets a demand floor on technology usage so that there is a market for it, slow and patient capital as well as loan guarantees will help to provide a more palatable marketplace for chip development and supply. But we've got to find some way to backstop the risk and to provide safety for society.
I think there are specific consumer protection aspects that we should be dealing with as well so if these products rise above what folks can afford how are we providing a safety net around them and some protections for their loss.
Speaker 39
Thank you very much for your answer. Yeah, no, absolutely. Mr. Chairman, I yield back.
Lori Trahan
Thank you very much, the gentleman yields back. The chair now recognizes the gentlelady from Massachusetts' third district for five minutes for questions.
Thank you, Mr. Chair. American leadership in AI is critical both in commercial frontier models and in open weight models. Most people are familiar with proprietary AI systems, models that you access through a company's products on their terms at their price. Open weight models are different.
When a company releases an open-weight model, anyone can take it, adapt it, and build on it, the same way open-source software has served as the bedrock of modern software development, unlocking rapid innovation. Open-weight models have a potential to do the same for AI. Just as open-source software can be modified by anyone to best fit their needs, open-weight models can be fine-tuned by anyone to be fit for the task at hand. Dr. Isler, how do open-weight models enable innovation for companies using AI, particularly smaller players?
Jedidiah Isler
Thank you so much for your question, Representative Trahan, and for your leadership and collaboration with FAF and AI privacy and more. We really appreciate the opportunity to work with you and your efforts. So, open-weight models, particularly those where not necessarily the data that it's trained on or the model itself are available, represent a reasonable opportunity for expanding the playing field in the ways that you talk about. I think it's important, though, to remember that the open-source or open-weight models are no more safe or less safe than proprietary models, so we still have the same considerations in terms of their use in malicious intent. So I wanted to just acknowledge that that's true.
However, by allowing pathways for reduced-cost engagement to these technologies, it does level the playing field and allow for an opportunity for small businesses, for universities, for nonprofits, those who may not be able to keep up in the marketplace with our proprietary colleagues.
Lori Trahan
Thank you. It is clear that open-weight models are critical for businesses back home and across the country to compete in this economy. So we need a robust ecosystem of American-made open models. But we are losing ground. Meta's Llama model was previously a leader in the open-weight space, but it's now been overtaken by models from Chinese companies like DeepSeek.
These Chinese models can be over ten times cheaper than comparable commercial models, yet deliver comparable performance. There are proposals to bring America back to the forefront of the open-weight space. Mr. Evans' draft bill, the Open-Weight AI Leadership Act, would task the Secretary of Commerce with increasing adoption of American open models. Similarly, my discussion draft with Mr. Obernolte provides smaller developers with the resources that they actually need to compete, such as compute, training data, and evaluation systems through the National AI Research Resource. Mr. Chilson, what further steps should Congress take to ensure that American open-model developers are sufficiently resourced with compute, training data, and other necessary resources to compete with China?
Neil Chilson
Well, thank you for your question, and thank you, Rep. Trahan, for your leadership on the Great American AI Act and all the work that you're doing in the tech policy space. So, there's, you already mentioned some of the proposals that are out there. I like to think of this on the supply and on the demand side. And so on the supply side, you know, the key inputs to model training are still expensive: data and energy. And anything that we can do to get those down will help, will help us develop more open models.
One thing that the government could do would be as a holder of massive amounts of useful data, including much that has nothing to do with personally identifiable information, the government should open that up to training, make it more available. And so I think that might be one input-side thing to do. On the energy side, there's obviously lots of work to be done on permitting and expanding. And then on the demand side, we could think of things supplying that demand, reserving federal compute for open market or open development, open-model development.
On the regulatory side, you know, a lot of the concerns that focus on licensing are going to be very challenging for open model providers to meet. Focusing more on narrow incident reporting is probably a better approach in that space and would incentivize people to develop open models instead of closed ones when they might be trying to make a decision.
Speaker 48
Great. Thank you for that. I do have a question for Mr. Jerbin about the governance and the ad hoc approach from this White House, but I will submit it for the record. Thank you. I yield back.
Speaker 39
Thank you very much. The gentlelady yields back and the chair now recognizes the gentlelady from New York's ninth district for five minutes for questions.
Speaker 49
Thank you very much, Mr. Chairman, and thank you Ranking Member Schakowsky, and good afternoon to our panelists. Thank you for joining us today. I'd be remiss if I did not acknowledge that yesterday the conservative Supreme Court majority ruled that Donald Trump can fire leaders of independent agencies, including the FTC, ending 90 years of court precedent to justify a firing based on ideologies that conflicted with the president. The firing of the Democratic commissioners of the FTC hinders the very agency set out to protect Americans from online harm and ensure a fair and competitive market for all consumers. This baseless firing continues the concerning pattern of systemic dismantling of any agency, grant program, or research institution that the president fears he can't control.
America loses when Trump does this. The slashing of billions of dollars in critical research funding has hindered our nation's ability to lead globally in research and development. It has forced research talent to opt out of working in the United States, has directly undermined public trust in government, and is already hurting our GDP. American global economic competitiveness requires investments in research and our workforce, not the chronic attack on top talent and research funding that we've seen in the last year and a half. The bipartisan CHIPS and Science Act is yet another example of the Trump's attempt to sabotage America's competitiveness.
The CHIPS and Science Act invested almost 53 billion dollars in the domestic semiconductor supply chain, including 11 billion for research and development. Since it became law, private entities have invested 640 billion in semiconductor manufacturing. The investments in this law are estimated to create and support about 500,000 jobs. So why, in the first three months in office, did Trump lay off 40 employees at the CHIPS Program Office? When we invest in ourselves, we undoubtedly see success.
We will not become global leaders if we divert our priorities from research and investment and towards unilateral and unregulated industry power. Innovation without safety is reckless. And when it comes to the people's health and safety, we cannot take the mindset of failing fast and often. It is Congress's responsibility to have more foresight and intention in protecting consumers while promoting American interests. I can guarantee you that cutting red tape and circumventing protections for the sake of protecting business interests simply lines the pockets of industry and does not help protect consumers and drive American success.
Being a global technology leader means nothing if you're not protecting the health and well-being of Americans. Dr. Eisler, what supply chain and environmental risks exist without substantive regulation of innovative technologies?
Jedidiah Isler
Thank you so much for the question.
It is true that the designation as a global tech leader does ring hollow if the cost for that leadership is the health and well-being of the American people. We view the reason to aspire to a leading position in science and technology leadership to be that it benefits all of society. It raises our economic mobility, it raises our GDP. So when we are technology leaders, we also are providing resource for our constituencies. This means that like if you want your community to thrive, you have to have an S&T strategy that accounts for the concerns you have around environmental concerns, supply chain concerns, and also pathways for folks to reap the benefits.
So I want to acknowledge that. We've also seen some of that progress reversed in the last few years, as you recall. We think that the best way to do this is responsible regulation, that you need to have guardrails that provide reasonable common sense safeguarding around industry and innovation, and also that allow room to have the sandbox to figure out what the next and best thing is. So we think you can have both, we think you have to consider both, and that when you do, you actually get a better product that allows us to be more globally competitive.
Speaker 49
Very well. I can appreciate the intention behind today's hearing, and many of the bills do address important technologies for American leadership. But we cannot overlook the historic disinvestment in research and government programs, or lose sight of the important regulations that keep Americans safe in this technological revolution and beyond. And with that, Mr. Chairman, I yield back.
Speaker 39
Well, thank you very much, the gentlelady yields back the balance of her time. The chair now recognizes himself for five minutes for questions. And again, to our witnesses, thanks very much for appearing today. The future of auto manufacturing should be built at home. That's why I'm proud to support the Connected Vehicle Security Act, which would ban connected Chinese vehicles from flooding our markets.
But a ban is only the first step. Congress must block unfair imports, but it must also ensure that America is the best place to build the vehicles of tomorrow. This is why I've been working on the SELF DRIVE Act. It would establish a federal framework for autonomous vehicles, giving automakers the regulatory certainty to build and deploy AVs that are as safe or safer than any vehicles on the road today. Mr. Durbin, I thank the Chamber of Commerce's strong support for the SELF DRIVE Act.
And Mr. Durbin, why is it important to have a federal AV framework to compete with China?
Marty Durbin
I'm sorry, sure, a federal framework for AVs. Well, as with almost any of the technologies we've discussed here, having that federal framework is what provides the consistency and the clarity on the investments that need to be made from the private sector in partnership with their government partners. So I think it is going to allow us to compete much more clearly against other nations.
Speaker 39
Well, let me follow up. If we do have, and good Lord willing, that the SELF DRIVE Act is going to pass this Congress, how would the SELF DRIVE Act impact American workers?
Marty Durbin
Chairman, if I could, I don't have enough of the details myself. Please let me get back to you in writing.
Speaker 39
Well, I'd appreciate that. You know, as we look at the SELF DRIVE Act especially, that unfortunately we've had about 42 to 43,000 people being killed on our highways every year. That 94% of all accidents that are caused out there are driver self, that they're not being attentive at the wheel.
So it's really important that we get this piece of legislation across the board, but if you could follow up with me and the committee, I'd greatly appreciate it. Dr. Eisler, in addition to robotics, China has made bio-manufacturing a national priority across successive five-year plans. That's why I'm leading the Bio-manufacturing Excellence, Domestic Resilience, Output, and Competitive Know-how, or BEDROCK Act. This legislation promotes the commercialization of bio-manufacturing, identify best practices, and shed light on federal approval processes. Dr. Eisler, is the United States treating bio-manufacturing with the same seriousness as our Chinese adversaries?
Jedidiah Isler
Thank you for the question. I think there remains significant opportunity across the US to really realize the potential that bio-manufacture can have for us. It is, FAS has worked on this topic for quite some time, and we see extensive excitement in various jurisdictions across the country to reinvigorate their economies through bio-manufacturing and bio-economy. So I think there's lots of room for us here, that BEDROCK is a really good start in acknowledging the other side of the bio-domain, that there is promise and there's something to promote. So I think there's a lot of opportunity for us there.
Speaker 39
Well, let me just make sure, are we, is the United States leading, are we equal, are we behind the Communist Chinese?
Jedidiah Isler
In, according to some of the studies we've seen in a number of these critical technologies and domains, we are actually behind on our investments and on our production.
Speaker 39
So, and again, just real briefly, how do we catch up and do we, you know, do we have the technology out there, are we not using it, or what's happening? Is it, is it regulations, is it laws?
Jedidiah Isler
I think it's investment in some of these early-stage technologies. There's lots of research on things like biofuels and the like that could benefit from additional attention and investment. We find the landscape for regulation to be one that depends on the context and what each region is interested in doing. And what we've really seen a need for is support for coordination across the nation. So if the federal government could help set a strategy on what we want to do in terms of bio-manufacturing, I think we'd find many people across the country at the ready to get into step with it so that we can make up some ground.
Speaker 50
Well, thank you very much. My time is about to expire and I will yield back the balance of my time. The chair now recognizes the gentleman from Georgia's first district for five minutes for questions.
Speaker 51
Well, thank you very much for being here. Thank you, Mr. Chairman, for this meeting. I want to kind of shift gears a little bit. I have the honor and privilege of representing the first congressional district of Georgia, which includes the entire coast of Georgia and includes two major seaports: the Port of Savannah, the number three container port in the country and the fastest-growing container port in the country, and the Port of Brunswick, the number one roll-on, roll-off port in the country. And logistics and supply chain hubs are extremely important, obviously, in our district and in our country, but especially in my district.
So when we talk about AI and we talk about robotics and we talk about cybersecurity and American competitiveness, it's not theoretical, it's real. It's real for us, it's real for us in the first district. Mr. Ferrello, I want to ask you, your testimony stated that advanced mobile robots are becoming common in industrial logistics, manufacturing, public safety, and all kind of defense and security applications. For a major logistics hub like the Port of Savannah, where do you see the most immediate value from trusted American robotics, like cargo movement, warehouse operations, infrastructure inspection, perimeter security, worker safety, or something else?
Jason Fiorillo
So thank you for the question. I would say all of the above. I'd say that robotics has tremendous potential to address many pressing needs in logistics hubs and logistics ports such as yours. We have a robot, for example, Boston Dynamics actually, that moves case goods eight hours a day, well, actually two shifts, 16 hours a day, 50-pound containers, one at a time, case goods, and we can unload 18-wheelers all day long. Our humanoid robotics actually will be enabled to do all sorts of...
Jason Fiorillo
Resembling a human in shape or appearance. So Atlas is our premier humanoid robot, and so it can pick things up, move things, operate things in environments that were built around the human form. And so our humanoid robot Atlas can do all sorts of different behaviors, and we are working on it to do such things, especially in logistics hubs.
Speaker 51
Should Congress be concerned about foreign-made robotics like this and our sensors being embedded in sensitive infrastructure like this?
Jason Fiorillo
That's a really great question, and thank you for asking it. Certainly as a domestic robotics company, that's something that we're concerned about as well. And so I think it's vital that trusted robots have trusted components and we have a trusted supply chain that is either domestic or friend-shored. Boston Dynamics wants to be part of the solution. But with respect to the other part of your question, there's a huge volume of robots that are being manufactured out of China, and we've seen many security vulnerabilities, cyber and security vulnerabilities where someone who is remotely situated either at the company that created it or possibly at the government can not only see what the robot is seeing, but can engage with or instruct the robot to take actions that perhaps are inconsistent with what the purchaser had hoped the robot would do.
Speaker 51
You know, I'm trying to keep myself from asking this question, but I can't. I just have to ask because, you know, we've done well on the East Coast with the relationship with the longshoremen and with the workers. West Coast not so much. In fact, we've gained a lot of business from them because of the problems they've had out there with the labor. I can imagine this caused a lot of problems with labor too.
Jason Fiorillo
Well, I think that anytime you introduce a new technology, there are changes that have to be made. What we've seen is when we introduce our technology into facility, it's actually highly welcomed. We've seen unions that actually bake cupcakes because they are so excited they no longer have to unload these 18-wheelers full of boxes and instead get upskilled and get better jobs within a logistics facility. So we think that in addition to that, it's vitally important for all of these workers that the manufacturing sectors that we have in the United States remain competitive by using advanced technology so that the centers themselves continue.
Speaker 51
Right. Mr. Durman, real quick, should a national robotics strategy specifically examine critical infrastructure like cases with ports and freight movement, warehouses, etc.?
Marty Durbin
No, I think any commission here should spot check those types of things to make sure that we can have some good recommendations back to policymakers on where to go from there.
Speaker 51
You know, the Georgia ports are the economic engine not only of Southeast Georgia, but of the Southeastern United States. The impact of the ports reaches all the way up into the Ohio Valley. So this is going to have a big, big impact on our ports and in a positive way and I hope in a positive way. So thank y'all very much. This is very important.
Thank you.
Speaker 39
Thank you very much. The gentleman yields back the balance of his time. The chair now recognizes the gentleman from Pennsylvania's 13th District for five minutes for questions.
John Joyce
Thank you, Mr. Chairman, for holding this important hearing. The topic of this hearing comes at a unique moment, the semiquincentennial of the United States.
This is a time to reflect on how far we've come as a nation over the past 250 years and to look ahead at to where we're still going. The United States leads on the world stage when it comes to technology. We have cemented our status among top nations in technological advancement and have achieved a level of global competitiveness that will allow us to secure our leadership in this critical space for decades to come. As we approach this 250 landmark, we must continue to evaluate ways to foster innovation and enhance the opportunities that we have been able to explore. We cannot grow complacent.
We must keep innovating, we must keep building, and we must keep competing. Mr. Durbin, this Congress, I introduced the Secure Data Act. This comprehensive federal privacy legislation that will protect consumers, that will enable beneficial uses of data, and give businesses the certainty that they need. Finding a path forward on federal privacy reform will end the state patchwork that currently stifles innovation and creates uncertain pathways for emerging technologies when it comes to data. Personal data plays a part in how technology is commercialized, including the technology discussed during the hearing today, such as alternative reality, robotics, and open source AI.
Mr. Durbin, how does a comprehensive federal privacy framework like the Secure Data Act support the commercialization of new technology?
Marty Durbin
Well, Mr. Joyce, thank you for the question and thank you for your leadership on the Secure Data Act as well. We appreciate that it does provide that that national framework. And as as noted in in my testimony, when we've got more than more than 20 states that have that have gone in separate directions on on privacy rules and AI rules, that doesn't help any of the companies involved in here.
So that is a patchwork of 20 states.
Correct.
Marty Durbin
The some of the the startup companies, the small businesses that are that are engaged in in the marketplace here have a very difficult time having to comply with different sets of rules in each state.
John Joyce
So without a federal framework, we are truly stifling the innovation of the small companies, of the entrepreneurs.
Marty Durbin
Yes, stifling the innovation and and loading enormous costs on top of them as well. I mean, right now, the California law itself estimates that each small business ends up paying $16,000 additional each year.
John Joyce
As a doctor who represents rural Pennsylvania in Congress, I know the importance of medical providers having access to cutting-edge technology that improves health and well-being for all of my patients and all of my constituents. Modern technology has truly revolutionized healthcare. From MRI imaging with PET and CT scans to ultrasound diagnostics, wearable cardiac monitors, these medical tools all rely on chips. Common devices such as pacemakers, defibrillators, cochlear implants, and portable diagnostic tools such as glucose monitors and insulin pumps are powered by semiconductors. Mr. Chilson, these advanced medical devices rely on semiconductor memory chips, and recent shortages have exposed the vulnerabilities in supply chains.
What can we do to address the memory chip shortage to ensure that our healthcare sector has the access to these critical technologies?
Neil Chilson
So, thank you so much for the question. I also share your optimism about the role of technology in healthcare sector. It's one of the best applications, one of the most promising applications of AI. It is an incredible tool in the doctor's tool belt. And the chip issue that you're raising, I think it is key for Congress and the bill that's at issue here could dig into this.
As I mentioned earlier, the memory space is complicated and not every, not all the chips that you're talking about that are in medical devices are the ones that are being, that have huge demand because of data centers. They might be affected by that demand, but it's not a one-to-one substitution. And so I think there has to be, as part of this inquiry that Congress and that the draft discussion bill talks about, we should examine how does this affect specific sectors like the medical sector, like the medical device sector, which is so promising and which we certainly wouldn't want a constrained supply of chips to slow down.
John Joyce
Could you imagine 250 years ago in Philadelphia where the framers of the Declaration of Independence and the United States Constitution were meeting in the hot summer, which included farmers and businessmen and yes, four doctors were part of that equation. These 250 years have shown that America is always looking to improve and to maintain that competitive edge. Securing US technology leadership right now will ensure that we can continue to invest, to innovate and invent for another 250 years. Thank you, Mr. Chairman, for allowing me to participate today and I yield the remainder of my time.
Speaker 55
The gentleman's time has expired and yields back. The chair now recognizes the gentlelady from Iowa's first district for five minutes for questions.
Mariannette Miller-Meeks
Thank you, Mr. Chairman. Given my colleague's comments, also as a doctor, I couldn't agree more about the potential for technology in healthcare, both in affordability, prevention and management. Advanced memory chips, as we've already heard, are essential to America's economic competitiveness and technological leadership. And at one time, the United States was certainly at the forefront of all of that innovation. As demand continues to grow, supply constraints and geopolitical risks threaten not only the development of emerging technologies like augmented intelligence, but also many industries that depend on reliable access to advanced semiconductors.
In many cases, supply shortages have driven up costs for manufacturers and businesses, forcing some companies to raise prices on the products and services consumers rely on everyday. And we also saw those supply chain constraints during the pandemic. My bill, the Memory Chip Competitiveness Assessment Act, would direct the Department of Commerce to study the advanced memory marketplace, including supply constraints, affordability, the role of countries of concern and barriers to expanding domestic production. The goal is to provide Congress with the data needed to develop targeted policies that strengthen our supply chains, reduce reliance on foreign adversaries and ensure the United States remains the global leader in advanced memory technology. I look forward to hearing from you in the answers to my questions.
With that, Mr. Durbin, you mentioned permitting reform and energy infrastructure in your testimony. How much do those issues affect the ability to expand advanced memory manufacturing in the United States?
Marty Durbin
Well, first of all, thank you for the question, Congresswoman, and thank you for your leadership in putting this legislation together as well. Well, permitting reform, as we've seen, is having a serious impact across the entire economy. So, not just in energy, but in roads and bridges and water and airports and in the build-out of manufacturing facilities. So, when we're looking at the supply-demand imbalance we have with memory chips, part of the answer here is how do we increase supply? And that, so how do we help facilitate the build-out of those manufacturing facilities?
So, that's clearly has an impact across the board here.
Mariannette Miller-Meeks
I couldn't agree more, as one of those members working in a bipartisan, bicameral way to advance permitting reform. We can't build in America if we don't have permitting reform on the table. Commerce is instructed to consult manufacturers, equipment suppliers, and downstream consumers in the study proposed in this bill. Are there additional stakeholders you believe should be included to ensure Congress receives a complete picture of the landscape?
Marty Durbin
I haven't seen the specifics, but I do think of those you named, I think those are the people that need to be at the table and can help us to identify, you know, where particular choke points are and the right policy solutions. Because we also want to be clear that, you know, this is a, this is a market issue as well. And you know, there are other industries that say, you know, the cure to high prices is high prices. I know there's a lot more complications in this particular case, but I think those are the voices we need at the table to help identify where we can find solutions.
Mariannette Miller-Meeks
Thank you. Mr. Chilson, one purpose of this bill is to distinguish temporary market shortages from longer-term structural supply challenges. What indicators should Commerce use to determine whether today's shortages reflect a cyclical market adjustment or a more permanent imbalance driven by augmented intelligence demand?
Neil Chilson
So, I think there's a whole host of factors, but we could start with just how durable is the demand? Is this, and in what sectors? As I already pointed out, memory is a very broad set of technologies, and in some of those areas, we may just be part of the cycle that's often in the memory space where there's a glut and then there's a shortage.
Mariannette Miller-Meeks
Thank you. Mr. Fiorillo, looking ahead to, given robotics, looking ahead to five to ten years, do you expect augmented intelligence-driven robotics to significantly increase demand for advanced memory chips, and should that projected demand influence US industrial policy today?
Jason Fiorillo
Alright, thank you for the question. For advanced mobile robotics, there are two different types of computes that we need to consider. First is the type of compute that is used to create the AI model, and so that would be in a typical data center, and so anytime you scale those, of course, you use more. The second would be within the robot itself. And so as the adoption and use of these robots hopefully grows in the United States, then certainly we would use more chips.
It's our hope, of course, that these chips continue to scale with the demand.
Speaker 39
Thank you very much. The gentleman yields back the balance of his time, and the chair seeing that there are no further members wishing to ask questions, this ends the questions from our members of the committee. Chair asks unanimous consent that the documents on the staff document list be submitted for the record. Without objection, so ordered. Chair thanks our witnesses for appearing before the subcommittee today.
Members may have additional written questions for you. Members are reminded that they have ten business days to submit questions for the record, and I, the chair also asks the witnesses to respond quickly to those questions. Members should submit their questions by the close of business on July the 14th, and without objection, the subcommittee is adjourned.