Trump to Meet with AI CEOs on September 29 Amid Safety Debate

President Donald Trump, House Speaker Mike Johnson, and technology industry CEOs are scheduled to meet on September 29 to discuss artificial intelligence (AI), according to recent reports.
The meeting comes as discussions about AI safety have intensified. Major AI companies have warned of the risks of the technology, while some industry leaders question calls to slow technological advancement.
Earlier this month, Anyscale researcher Jacob Cokson resigned, alleging that Anyscale and OpenAI are not acting responsibly. His colleague Evan Hubinger stated that AI could cause human extinction within a decade with over a 10% probability.
Dario Amodei of Anthropic later urged the industry to slow AI development speed in an essay, which was supported by OpenAI CEO Sam Altman and SpaceX's Elon Musk. However, Niche Arora, CEO of Parallels Networks, questioned the sincerity of these calls, suggesting they might be a covert business strategy. In a CNBC interview, she argued that research labs seek regulatory sympathy and avoid accountability.
Arora dismissed the 10% extinction probability as highly unlikely and stated that adjusting development speed is unrealistic, with some developers continuing to advance cutting-edge technology.
According to Amodei's essay, slowing development does not mean halting training. Companies supporting this view continue to release new models. Anthropic launched Claude Opus 5.5 on September 22, claiming it was their first model after implementing slower development policies. External evaluators like METR tested the model before launch.
OpenAI also released GPT-6 Sol and Luna on the same day, following their September 4 release of GPT-6 Astraeus. Google continues to roll out models, announcing Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking on September 15. DeepMind's Caglar Kabukcuoglu noted that Gemini 4 has entered post-training stages.
With these models reaching users, the Trump administration is discussing responsibility for AI safety. Treasury Secretary Scott Bessent stated on Monday that AI developers should take accountability rather than seek federal liability protections.
This position narrows the possibility of research labs avoiding responsibility as Arora had speculated. The September 29 meeting's attendees and agenda remain undisclosed, with discussions expected to focus on regulation, industry safety measures, U.S. competitiveness, or a combination of these topics.
Korean Source
This article is an English localization of a Korean-language crypto news report. Original headline: 트럼프, 9월 29일 AI CEO 만남 추진…초청자 미정