House bipartisan bill directs NSA to create 'AI security playbook' amid Chinese tech race

Fox News

2 days ago

FIRST ON FOX – Rep. Darin LaHood, R-Ind., is introducing a new bill Thursday imploring the National Security Administration (NSA) to develop an "AI security playbook" to stay ahead of threats from China and other foreign adversaries.
The bill, dubbed the "Advanced AI Security Readiness Act," directs the NSA's Artificial Intelligence Security Center to develop an "AI Security Playbook to address vulnerabilities, threat detection, cyber and physical security strategies, and contingency plans for highly sensitive AI systems."
It is co-sponsored by House Select Committee on China Chairman Rep. John Moolenaar, R-Mich., Ranking Member Rep. Raja Krishnamoorthi, D-Ill., and Rep. Josh Gottheimer, D-N.J.
LaHood, who sits on the House Intelligence Committee and the House Select Committee on China, told Fox News Digital that the legislative proposal, if passed, would be the first time Congress codifies a "multi-prong approach to ensure that the U.S. remains ahead in the advanced technology race against the CCP." He said the bill will improve export control mechanisms – including for chips and high capacity chip manufacturing – protect covered AI technologies with a focus on cybersecurity, and limit outbound investment to firms directly tied to the Chinese Community Party or China's People's Liberation Army.
"We start with the premise that China has a plan to replace the United States. And I don't say that to scare people or my constituents, but they have a plan to replace the United States, and they're working on it every single day. And that entails stealing data and infiltrating our systems," LaHood told Fox News Digital. "AI is the next frontier on that. We lead the world in technology. We lead the world when it comes to AI. But what this bill will do will again make sure that things are done the right way and the correct way, and that we're protecting our assets and promoting the current technology that we have in our country."
LaHood pointed to evidence uncovered by the committee that he said shows the CCP's DeepSeek used illegal distillation techniques to steal insights from U.S. AI models to accelerate their own technology development. He also pointed to how China allegedly smuggled AI chips through Singapore intermediaries to circumvent U.S. export controls on the technology.
"As we look at, 'How do we win the strategic competition?' I think most experts would say we're ahead in AI right now against China, but not by much. It is a short lead," LaHood told Fox News Digital.
He said he is confident the bill will put the U.S. "in the best position to protect our assets here and make sure that we're not shipping things that shouldn't go to AI that allow them to win the AI race in China."
"Whoever wins this race in the future, it's going to be critical to future warfare capabilities, to, obviously, cybersecurity," LaHood continued. "And then, whoever wins the AI competition is going to yield really unwavering economic influence in the future. And so we're aggressive in this bill in terms of targeting those areas where we need to protect our AI and our companies here in the United States, both on the commercial side and on the government side, to put us in the best position possible."
The legislative proposal calls on the NSA to develop a playbook that identifies vulnerabilities in AI data centers and developers producing sensitive AI technologies with an emphasis on unique "threat vectors" that do not typically arise, or are less severe, in the context of conventional information technology systems." The bill says the NSA must develop "core insights" in how advanced AI systems are being trained to identify potential interferences and must develop strategies to "detect, prevent and respond to cyber threats by threat actors targeting covered AI technologies."
The bill calls on the NSA to "identify levels of security, if any, that would require substantial involvement" by the U.S. government "in the development or oversight of highly advanced AI systems." It cites a "hypothetical initiative to build covered AI technology systems in a highly secure government environment" with certain protocols in place, such as personnel vetting and security clearance processes, to mitigate "insider threats."
Though not directly related, the legislation is being introduced a week after FBI Director Kash Patel sounded the alarm on how the CCP continues to deploy operatives and researchers to "infiltrate" U.S. institutions. Patel laid out the risk in announcing that two Chinese nationals were charged with smuggling a potential bioweapon into the U.S.
LaHood said that case further highlights "the level of penetration and sophistication that the CCP will engage in," but he added that his bill focuses on putting a "protective layer" on U.S. AI tech and "restricting outbound investment to China." He pointed to how the CCP also has bought up farmland around strategic U.S. national security locations, particularly in Montana, North Dakota and South Dakota.
"If everything was an even playing field, and we were all abiding by the same rules and standards and ethical guidelines, I have no doubt the U.S. would win [the AI race], but China has a tendency and a history of playing by a different set of rules and standards," LaHood said. "They cheat, they steal, they take our intellectual property. Not just my opinion, that's been factually laid out, you know, in many different instances. And that's the reason why we need to have a bill like this."
The bill comes as the Trump administration has been pushing to bolster artificial intelligence infrastructure in the United States, and major tech companies, including Amazon, Nvidia, Meta, OpenAI, Oracle and others, have made major investments in constructing AI-focused data centers and enhancing U.S. cloud computing. Last week, Amazon announced a $20 billion investment in constructing AI data centers in rural Pennsylvania. It followed a similar $10 billion investment in North Carolina.
In late May, the NSA's Artificial Intelligence Security Center released "joint guidance" on the "risks and best practices in AI data security." The recommendations include implementing methods to secure the data used in AI-based systems, "such as employing digital signatures to authenticate trusted revisions, tracking data provenance, and leveraging trusted infrastructure." The center said its guidance is "critically relevant for organizations – especially system owners and administrators within the Department of Defense, National Security Systems, and the Defense Industrial Base – that already use AI systems in their day-to-day operations and those that are seeking to integrate AI into their infrastructure."