Elizabeth Warren accuses AI firms of using trade accord to dodge regulatory oversight

Warren accuses AI firms of trying to curb oversight in trade accord

WASHINGTON, July 23 (Reuters) - Democratic Senator Elizabeth Warren accused major AI companies of pushing the Trump administration to limit AI transparency rules in North America's trade pact, arguing instead that regulators should have broader access to information about the companies' models. Warren, the top Democrat on the Senate Banking Committee, said the companies are lobbying for negotiators to weaken the ability of regulators to oversee AI companies via the U.S.-Mexico-Canada Agreement. The agreement currently only requires disclosure of AI model details during government investigations. In a letter dated Wednesday, Warren urged the U.S. Trade Representative Jamieson Greer to resist those calls and to instead strengthen requirements that would give regulators across the region access to key information about AI models even absent a formal probe. "Big Tech ⁠is advocating to update the USMCA to strengthen provisions that allow companies to hide their AI algorithms and source code from regulatory scrutiny, even as the potential risks that AI poses to American families flow across borders and the need to regulate AI becomes increasingly evident," she wrote. Any revised agreement "should eliminate source code secrecy provisions that allow Big Tech companies to escape regulatory scrutiny," she said. "It is crucial that governments have the tools they need to effectively protect their citizens against these harms," she added. The USTR did not immediately respond to a request for comment. The letter, not previously reported, is part of a growing chorus of lawmakers on both sides of the aisle calling for tougher AI regulation following a fresh ⁠round of examples of AI causing harm this week. The New York Times reported on Wednesday, opens new tab that a man has sued OpenAI after its AI chatbot directed him to ignore dangerous health symptoms. OpenAI said on Tuesday that an autonomous agent powered by its advanced AI models went rogue during a security test and triggered a hack that compromised the infrastructure of AI startup Hugging Face. The ⁠incident "underscores the critical need for clear, practical rules for advanced AI systems," Republican Congressman Jay Obernolte said. Industry trade groups have pressed the USTR to seek looser digital trade rules from America's trading partners. In a comment to the trade authority last October, the Computer & Communications ⁠Industry Association, which represents AI giants Google and Meta described "forced disclosure of source code, algorithms, model weights and training data;" as "threats to the cross-border supply of AI services." In a separate comment to USTR that same month, the Telecoms Industry ⁠Association, which represents Apple and Amazon Web Services, said, "it is more important than ever that USTR push back against efforts by China and others to mandate source code disclosure." The groups did not immediately respond to requests for comment. Reporting by Alexandra Alper; Additional reporting by Courtney Rozen and David Lawder in Washington; Editing by Kim Coghill Our Standards: The Thomson Reuters Trust Principles., opens new tab

Warren accuses AI firms of using trade deals to escape oversight

The senator says the industry is converting contested AI rules into trade concessions, where they face far less public scrutiny. Senator Elizabeth Warren has accused the biggest AI companies of trying to write their own regulatory ceiling into a trade agreement, using the machinery of US trade policy to lock in limits on oversight that Congress has so far declined to grant them. The accusation, reported by Reuters on July 23, extends a fight the Massachusetts Democrat has waged for more than a year. Her argument is that Big Tech is quietly turning contested domestic questions into trade concessions, the same logic behind the trade probe into EU tech rules that lawmakers pushed the administration to open. The mechanism she points to is a clause, not a headline. A run of recent US bilateral deals, catalogued in a Center for Strategic and International Studies analysis of what it calls a "containment doctrine", carry near-identical "non-discrimination" language that bars partners from measures that "discriminate against U.S. digital services". That wording is broad enough to reach the rules the industry has spent two years trying to blunt: the EU's Digital Markets Act and Digital Services Act, and the patchwork of state AI statutes that Washington has separately tried to preempt at home. A rule that survives a state legislature can still be challenged as a trade barrier abroad. Warren has made versions of this case straight to the administration. In a May 2025 letter to Treasury Secretary Scott Bessent, Commerce Secretary Howard Lutnick, and trade representative Jamieson Greer, she wrote that "Big Tech firms have long sought to use trade deals to undermine pro-consumer, pro-competition policies". That letter named Apple, Amazon, Meta, and Google as funders of the industry groups pressing the US Trade Representative to treat foreign digital rules as barriers. A follow-up in March 2026, prompted by the volume of sexual deepfakes generated with Elon Musk's Grok, asked whether US negotiators had leaned on other countries to drop content-moderation laws outright. In that letter she pointed to the administration's handling of the EU's Digital Services Act, which US officials have described as "overseas extortion", alongside stalled talks over the UK's Online Safety Act and France's SREN law, which stiffens penalties for sexual deepfakes. The through-line, she argued, is that content and competition rules keep surfacing at the negotiating table rather than in open rulemaking, where they can be debated and voted on. The firms' side of this is less a rebuttal than a reframing. The industry has argued consistently that measures like the DMA are protectionism dressed as consumer protection, discriminatory by design and therefore fair game for a trade complaint, an argument the same companies made while they bankrolled the campaign against AI rules at home. The administration has been willing to back that reading with leverage, at one point threatening 100% tariffs over digital services taxes. The CSIS analysis notes the tactic works unevenly: smaller economies such as Guatemala and Cambodia have signed accords carrying the restrictive clauses, while larger markets have held out. The domestic side of the fight runs in parallel. Washington has separately weighed a federal standard that would override state AI statutes, and the industry has spent months girding to block those state rules, so a trade clause that reaches the same laws from outside the country would hand the effort a second lever. What Warren cannot easily do is force the text into daylight. Trade negotiations are conducted with limited transparency, and she has co-sponsored legislation to require that deal text be published before Congress grants fast-track authority, so far without success. Without the text, she has argued, neither lawmakers nor the public can judge what has actually been promised. Whether the July charge lands any differently is unclear. Reuters reported the accusation; the companies named in her earlier letters have not addressed it publicly, and USTR has not answered the specific claim. For now it is one senator's reading of a clause most voters will never see, which is close to the point she is making.

Warren accuses AI firms of trying to curb oversight in trade accord

Senator Elizabeth Warren criticizes major AI firms for lobbying against transparency rules. These companies want to hide AI algorithms and source code from regulators. Warren urges the U.S. Trade Representative to strengthen oversight requirements for AI models. Recent AI incidents highlight the urgent need for clearer regulatory frameworks. This push for stricter rules comes amid growing concerns about AI's potential harms. Democratic Senator Elizabeth Warren accused major AI companies of pushing the Trump administration to limit AI transparency rules in North America's trade pact, arguing instead that regulators should have broader access to information about the companies' models. Warren, the top Democrat on the Senate Banking Committee, said the companies are lobbying for negotiators to weaken the ability of regulators to oversee AI companies via the U.S.-Mexico-Canada Agreement. The agreement currently only requires disclosure of AI model details during government ⁠investigations. In a ⁠letter dated Wednesday, Warren urged the U.S. Trade Representative Jamieson Greer to resist those calls and to instead strengthen requirements that would give regulators across the region access to key information about AI models even absent a formal probe. "Big Tech is advocating to update the USMCA to strengthen provisions that allow companies to hide their AI algorithms and source code from regulatory scrutiny, even as the potential risks that AI poses to American families flow across borders and the need to regulate AI becomes increasingly evident," she ⁠wrote. Any revised agreement "should eliminate source code secrecy provisions that allow Big Tech companies to escape regulatory scrutiny," she said. "It is crucial that governments have the tools they need to effectively protect their ⁠citizens against these harms," she added. The USTR did not immediately respond to a request for comment. The letter, not previously reported, is part of a growing chorus of lawmakers on both sides of the aisle calling for tougher AI regulation following a fresh round of examples of AI causing harm this week. The New York Times reported on Wednesday that a man has sued OpenAI after its AI chatbot directed him to ignore dangerous health symptoms. OpenAI said on Tuesday that an autonomous agent powered by its advanced AI models went rogue during a security test and triggered a hack that compromised the infrastructure of AI startup Hugging Face. The incident "underscores the critical need for clear, practical rules for advanced AI systems," Republican Congressman Jay Obernolte said. Industry trade groups have pressed the USTR ⁠to seek looser digital trade rules from America's trading partners. In a comment to the trade authority last October, the Computer undefined" as "threats to the cross-border supply of AI services." In a separate comment to USTR that same month, the Telecoms Industry Association, which represents Apple and Amazon Web Services, said, "it is more important than ever that USTR push back against efforts by China and others to mandate source code disclosure." The groups did not immediately respond to requests for comment.

Warren accuses AI firms of trying to curb oversight in trade accord

WASHINGTON, July 23 (Reuters) - Democratic Senator Elizabeth Warren accused major AI companies of pushing the Trump administration to limit AI transparency rules in North America's trade pact, arguing instead that regulators should have broader access to information about the companies' models. Warren, the top Democrat on the Senate Banking Committee, said the companies are lobbying for negotiators to weaken the ability of regulators to oversee AI companies via the U.S.-Mexico-Canada Agreement. The agreement currently only requires disclosure of AI model details during government investigations. In a letter dated Wednesday, Warren urged the U.S. Trade Representative Jamieson Greer to resist those calls and to instead strengthen requirements that would give regulators across the region access to key information about AI models even absent a formal probe. "Big Tech is advocating to update the USMCA to strengthen provisions that allow companies to hide their AI algorithms and source code from regulatory scrutiny, even as the potential risks that AI poses to American families flow across borders and the need to regulate AI becomes increasingly evident," she wrote. Any revised agreement "should eliminate source code secrecy provisions that allow Big Tech companies to escape regulatory scrutiny," she said. "It is crucial that governments have the tools they need to effectively protect their citizens against these harms," she added. The USTR did not immediately respond to a request for comment. The letter, not previously reported, is part of a growing chorus of lawmakers on both sides of the aisle calling for tougher AI regulation following a fresh round of examples of AI causing harm this week. The New York Times reported on Wednesday that a man has sued OpenAI after its AI chatbot directed him to ignore dangerous health symptoms. OpenAI said on Tuesday that an autonomous agent powered by its advanced AI models went rogue during a security test and triggered a hack that compromised the infrastructure of ?AI startup Hugging Face. The incident "underscores the critical need for clear, practical rules for advanced AI systems," Republican Congressman Jay Obernolte said. Industry trade groups have pressed the USTR to seek looser digital trade rules from America's trading partners. In a comment to the trade authority last October, the Computer & Communications Industry Association, which represents AI giants Google and Meta described "forced disclosure of source code, algorithms, model weights and training data;" as "threats to the cross-border supply of AI services." In a separate comment to USTR that same month, the Telecoms Industry Association, which represents Apple and Amazon Web Services, said, "it is more important than ever that USTR push back against efforts by China and others to mandate source code disclosure." The groups did not immediately respond to requests for comment. (Reporting by Alexandra Alper; Additional reporting by Courtney Rozen and David Lawder in Washington; Editing by Kim Coghill)