Sam Altman to meet with Trump administration amid China AI debate
WASHINGTON, DC – JANUARY 21: OpenAI CEO Sam Altman appears at a press conference with US President Donald Trump in the Roosevelt Room of the White House on January 21, 2025 in Washington DC. Trump announced he would invest in artificial intelligence (AI) infrastructure and took questions on January 6 on a variety of topics, including presidential pardons for criminal defendants, the war in Ukraine, cryptocurrencies and other topics. (Photo: Andrew Harnik/Getty Images)
Andrew Harnik | Getty Images News | Getty Images
OpenAI CEO Sam Altman will meet with senior Trump administration officials, lawmakers and economists in Washington, D.C. this week to preview the capabilities of the company’s upcoming family of AI models, CNBC has learned.
Altman has maintained a regular presence on Capitol Hill this year to help ease policymakers’ concerns about rapidly advancing technology and shape their views on what regulations should look like.
The visit comes at a particularly sensitive time for AI development in the United States, where debate over whether to restrict China’s vulnerability-heavy AI models is sweeping Silicon Valley.
Altman is expected to preview future releases of OpenAI as well as answer questions from policymakers about cybersecurity and the company’s stance on vulnerability-heavy models, according to a source familiar with the plans who asked not to be named because the details are confidential.
Vulnerable AI models are available for users to download, modify, and run their own infrastructures. In recent weeks, Chinese vulnerability-heavy models from startups like Moonshot AI have quickly caught up with leading, proprietary offerings from American companies including OpenAI and its chief rival Anthropic, sparking concern about the durability of the U.S. lead in the AI race.
Officials and executives are debating whether to limit access to Chinese models in the United States, prompting a coalition of tech companies including Nvidia, Microsoft, Meta and Palantir to issue a letter on Friday urging policymakers to avoid “premature restrictions” on deficit-heavy models.
OpenAI signed the letter after it was first published, but the company’s allies lobbied for restrictions on open-heavy models. New York Times. Antropik has not yet signed or made any official comments on its position.
Altman will also likely touch on the “unprecedented cyber incident” that OpenAI announced earlier this month, which shook researchers and executives in the tech industry.
The company said its models escaped its sandbox testing environment, accessed the internet, and exploited a vulnerability to break into another company’s systems. OpenAI said it tried to find information that models could use to cheat on an evaluation, and they were successful.
The breached company, Hugging Face, operates an open-source developer platform. Hugging Face shared An autonomous AI agent system breached part of the production infrastructure before OpenAI announced its involvement.
AI companies including OpenAI have been warning about the technology’s advancing cyber capabilities, especially since Anthropic launched a model called the Claude Mythos Preview in April. Following the incident, OpenAI said that AI is accelerating the discovery and exploitation of vulnerabilities, which means model safety and security must keep pace.
The company said it was working to strengthen “containment, monitoring, access controls, and evaluation practices used during model development” and was conducting a “thorough investigation” into the incident. a version.
The hacking incident demonstrated how powerful and resourceful autonomous AI agents have become.
While in D.C., Altman is also expected to discuss the evolution of AI agents and their impact on worker productivity, the person said. He will explain the concept of “AI teams,” where multiple agents work together to accomplish long-term tasks.
WRISTWATCH: OpenAI model goes bad: Here’s what you need to know
