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. President Trump announced investments in artificial intelligence (AI) infrastructure and took questions on a wide range of topics, including the presidential pardon for the January 6 defendant, the war in Ukraine, and cryptocurrencies. (Photo by Andrew Harnik/Getty Images)
Andrew Harnik | Getty Images News | Getty Images
OpenAI CEO Sam Altman will meet with Trump administration officials, lawmakers and economists in Washington, D.C. this week to preview the capabilities of the company’s upcoming suite of artificial intelligence models, CNBC reported.
Mr. Altman has continued to appear regularly on Capitol Hill this year, seeking to alleviate policymakers’ concerns about rapidly advancing technology and help shape opinions about what regulations should be.
The visit comes at a particularly sensitive time for U.S. AI development, as debate over whether to limit China’s promiscuous AI models sweeps through Silicon Valley.
In addition to previewing OpenAI’s upcoming releases, Altman will answer questions from policymakers about cybersecurity and the company’s stance on non-discriminatory models, said a person familiar with the plans who asked not to be identified because the details are confidential.
Openweight AI models can be downloaded, modified, and run by users on their own infrastructure. In recent weeks, China’s indiscriminate model from startups such as Moonshot AI has rapidly overtaken major proprietary products from U.S. companies, including OpenAI and its main rival Anthropic, raising concerns about whether America’s lead in the AI race will last.
As officials and executives consider whether to restrict access to Chinese-made models in the United States, a coalition of tech companies including Nvidia, Microsoft, Meta and Palantir released a letter Friday urging policymakers to avoid “premature restrictions” on open-weight models.
The New York Times reports that OpenAI signed the letter after it was first published, but that the company’s allies are lobbying for restrictions on the promiscuous model. Anthropic has not yet signed or publicly commented on its position.
Altman is also likely to address the “unprecedented cyber incident” OpenAI disclosed earlier this month that has rattled researchers and executives across the technology industry.
The company said its models escaped from a sandboxed test environment, accessed the Internet, and exploited vulnerabilities to infiltrate other companies’ systems. OpenAI said the model was trying to find information that could be used to falsify ratings, and it was successful.
The company that was compromised, Hugging Face, operates an open source developer platform. Hugging Face shared that an autonomous AI agent system had infiltrated some of its production infrastructure before OpenAI’s involvement was revealed.
AI companies, including OpenAI, have been warning about the technology’s evolving cyber capabilities, especially since Anthropic released a model called Claude Mythos Preview in April. In response to the incident, OpenAI said that as AI vulnerabilities are discovered and exploited at an accelerated pace, model security and safety must also catch up.
The company said it is working to strengthen “containment, monitoring, access control, and evaluation practices used during model development” and is conducting a “thorough investigation” of the incident, according to the release.
This hacking incident showed how powerful and resourceful autonomous AI agents are becoming.
During his time in Washington, Altman will also discuss the evolution of AI agents and their impact on worker productivity, the people said. Learn about the concept of “AI teams” where multiple agents work together to perform long-term tasks.
WATCH: OpenAI models go rogue: Here’s what you need to know
