OpenAI's Altman to Brief Trump Administration on New Models as U.S. Pushes for AI Strategy

OpenAI's Altman to Brief Trump Administration Ahead of New Model Launch

OpenAI CEO Sam Altman is heading to Washington next week to meet with the Trump Administration and U.S. lawmakers to discuss new artificial intelligence models, as the government looks to create a framework for reviewing AI systems. The discussions are expected to include OpenAI's upcoming models, the broader implications of advanced AI systems, and how they will affect the way people work, OpenAI's global public affairs chief, Chris Lehane, said at a briefing in Washington, Bloomberg reported. "We think that there's going to be some really interesting capabilities with this model family, particularly as it relates to work and scaling work," Lehane said. The Trump administration is considering creating a standalone oversight body to evaluate AI systems. One version of the proposal, developed with input from Treasury Secretary Scott Bessent, aims to give AI companies clearer expectations surrounding how these models would be assessed. The idea closely monitors a framework recently proposed by Google DeepMind CEO Demis Hassabis, who is set to meet with Washington offices this week to advocate for a similar approach, Bloomberg added. Lehane noted that if Congress can't create a national standard for AI governance, the plan would be to create a reverse federalism approach, where states would mirror each other regarding rules and regulations surrounding AI. OpenAI is backing a proposal in Massachusetts that would require large AI companies to do more to address the potential for severe AI-related risks. The initiative also received support from Anthropic. Lehane added that the U.S. needs a Coordinated national AI strategy as competition with China intensifies. His comments came after Chinese AI startup Moonshot launched a new model Kimi K3 that challenged deleting US systems from companies such as OpenAI and Anthropic, raising concerns about whether massive AI infrastructure investments will deliver expected returns. Lehane argued that a unified approach is needed to maintain US technological leadership, strength and cybersecurity defenses and ensure that AI models can be safely shared with US allies. He pointed to China's government-backed AI strategy as an example of the coordinated effort that the U.S. may need to compete globally. Last month, Anthropic had to abruptly disable access to its flagship Fable 5 and Mythos 5 artificial intelligence models after the U.S. government ordered the startup to block foreign nationals from using the systems, citing national security authorities. The company said authorities did not provide detailed justification for the directive. However, Anthropic said it understands the government's concerns related to a potential method of "jailbreaking" Fable 5 -- a technique used to bypass a model's built-in safeguards. Market News and Data brought to you by Benzinga APIs To add Benzinga News as your preferred source on Google, click here.

Sam Altman to Brief White House on OpenAI Models' Impact on Work | PYMNTS.com

The report cited remarks made at a briefing in Washington by OpenAI Chief Global Affairs Officer Chris Lehane. "We think that there's going to be some really interesting capabilities with this model family, particularly as it relates to work and scaling work," Lehane said, according to the report. Lehane also said during the briefing that the United States needs a more comprehensive AI strategy to compete with China and ensure cybersecurity specialists in the U.S. and its allies can access the latest models, per the report. He also said that until there is a national standard, OpenAI will work with states and encourage them to adopt a single approach to AI governance, according to the report. Lehane wrote in a Wednesday (July 15) blog post that California, New York and Illinois have advanced frontier safety legislation that could lay the foundation for a national standard and a U.S.-led global AI framework. The key elements of the states' legislation include a documented safety framework with risk assessments for frontier models, public disclosure of those assessments and their results, reporting of serious safety incidents, and governance and accountability through independent, objective audits, Lehane said. "In particular, and as the last few weeks have shown, as AI models become more and more capable, we need a coherent system that will allow us to get the tools into the hands of the government, critical infrastructure, allies and trusted partners," Lehane said in the post. It was reported Friday (July 17) that the White House is exploring plans to create a regulator focused on AI safety. The plan follows complaints from Silicon Valley about the government's recent curbs on new AI models from companies like Anthropic and OpenAI. On July 14, Google DeepMind CEO Demis Hassabis called for a U.S.-led standards body to independently test the world's most advanced AI models for national security threats before release. The body would test dangerous capabilities, cybersecurity risks and whether model safeguards can be bypassed.