AI News Roundup – The New York Times and Chicago Tribune sue Perplexity AI over copyright infringement, state-level AI preemption push stalls in US Congress, new US bill would strengthen export controls for AI chips, and more

To help you stay on top of the latest news, our AI practice group has compiled a roundup of the developments we are following.

• • The New York Times and the Chicago Tribune are suing Perplexity AI alleging copyright infringement, according to CNBC. The lawsuit, filed in federal court in New York, alleges that Perplexity illegally scraped the papers’ copyrighted web content for use by its AI models, which are alleged to generate outputs that are “identical or substantially similar to” the copyrighted content. This case is the latest in a series of copyright infringement lawsuits to be brought against Perplexity, including a suit by the publisher of The Wall Street Journal and New York Post in October 2024 and a suit by Reddit in October 2025. A spokesman from The Times said in a statement that “we firmly object to Perplexity’s unlicensed use of our content to develop and promote their products,” and that the company “will continue to work to hold companies accountable that refuse to recognize the value of our work.” The case is Civil Action No. 1:25-cv-10106 in the United States District Court for the Southern District of New York.
  • Axios reports that a push by Republican lawmakers in the U.S. Congress to preempt state-level AI regulations is floundering. Senator Ted Cruz of Texas and House Majority Leader Steve Scalise of Louisiana have proposed a measure to preempt state-level AI regulations without any additional federal AI rules to be included in this year’s annual defense funding bill, though sources told Axios that the AI measure will be unlikely to be included. Such preemption has received wide criticism from the opposition Democratic Party but also many Republican officials, including Spencer Cox, the governor of Utah, who has pitched a “pro-human AI” initiative aimed at protecting children and data privacy. In lieu of federal legislation, Republicans may attempt to fight state-level AI regulations through an executive order, as this AI Roundup reported last month, though no directive from the White House has yet appeared.
  • A new bill proposed in the U.S. Senate would strengthen export controls on the sale of advanced AI chips to China, according to Bloomberg. The Secure and Feasible Exports Act, or SAFE Act, would direct the Department of Commerce to cease issuing export licenses for chips to a list of adversaries, including China and Russia, for at least two and a half years. Nvidia, the market leader in advanced AI chips, has been lobbying against restrictions on exports, and recently scored a win by convincing lawmakers to strip out a restrictive export measure from an annual defense funding bill. However, the SAFE Act has bipartisan support, though the Trump administration has yet to comment on the measure. In response to the new bill, an Nvidia spokesman said that “nonmilitary businesses everywhere should be able to choose the American technology stack, promoting U.S. jobs and promoting national security.”
  • The Financial Times reports that the European Union is investigating Meta over antitrust concerns related to WhatsApp’s AI policies. Earlier this year, Meta integrated its Meta AI assistant into WhatsApp, one of the world’s largest messaging platforms, allowing users to ask questions or generate messages within chats. However, the European Commission has expressed concerns that Meta’s policy to prevent rival AI models from running on WhatsApp could violate EU antitrust law. Italy’s antitrust enforcement agency is already investigating Meta over its WhatsApp AI policies for allegedly using its dominant market position to integrate AI into WhatsApp without user consent. Meta has asserted that the Italian claims are “baseless” and that “the AI space is highly competitive.”
  • Estée Lauder and Google have partnered to develop a new AI chatbot designed to sell fragrances, according to The Wall Street Journal. Jo Malone London, one of Estée Lauder’s fragrance brands, now is making use of an “AI Scent Advisor” to help customers understand fragrances before buying them. The company partnered with engineers from Google Cloud, who fed scent categories and descriptions of the scents into Google’s Gemini AI model. The AI Scent Advisor then communicates with customers the same way an in-store associate would do so to find scents that align with customer wants. The move appears to have helped sales numbers, and Estée Lauder is considering bringing the AI Scent Advisor to its other fragrance brands.