The Trump administration is weighing the imposition of oversight on new A.I. models prior to their public release, reflecting a significant shift from its previously noninterventionist stance. U.S. officials have indicated that the administration is considering an executive order to establish an A.I. working group, collaborating with tech executives from companies like Anthropic, Google, and OpenAI, to explore potential review processes for A.I. safety. This discussion comes in response to concerns raised by the release of Anthropic’s recent model, and follows the administration’s earlier rollback of Biden’s regulatory efforts aimed at ensuring A.I. safety evaluations. The proposed oversight could be influenced by similar measures in the U.K., where multiple government bodies assess A.I. models for compliance with safety standards.

Google: Google, a core division of Alphabet Inc., leads in AI research through initiatives like DeepMind and Gemini models. Its executives were briefed by White House officials on plans for AI oversight including pre-publication reviews. Recent meetings reflect the company’s involvement in federal deliberations on AI governance.
OpenAI: OpenAI is a key AI research lab known for generative models that power applications like ChatGPT. Leaders from OpenAI participated in White House discussions about an upcoming executive order for AI model vetting. The firm is navigating evolving U.S. policy as the administration considers safety measures.
Dustin Volz: Dustin Volz is a journalist reporting from Washington on technology policy and cybersecurity issues. He collaborated on the New York Times piece revealing the administration’s plans for an AI working group to examine oversight on new models. His contributions focus on the intersection of government and tech sectors.
Tripp Mickle: Tripp Mickle is a New York Times reporter covering major technology companies such as Nvidia, Google, and Apple along with industry trends like artificial intelligence. He co-authored the article detailing the Trump administration’s discussions on imposing government oversight over new AI models. Based in San Francisco, his reporting highlights developments at the intersection of tech and policy.
Alphabet Inc.: Alphabet Inc. serves as the parent company of Google and invests across AI and other technologies. It is connected to the news through Google’s executives engaging with the White House on AI oversight proposals. Recent policy talks highlight Alphabet’s stake in federal AI regulation developments.
Sheera Frenkel: Sheera Frenkel is a New York Times reporter based in the San Francisco Bay Area focusing on how technology influences everyday lives and social media platforms. She contributed to coverage of White House meetings with AI executives from Anthropic, Google, and OpenAI regarding potential AI oversight procedures. Her work examines the societal impacts of emerging tech.
Anthropic AI LLC: Anthropic AI LLC develops AI systems with a strong emphasis on safety, recently introducing advanced models like Mythos that have drawn regulatory scrutiny. Executives from the company met with White House officials last week to discuss potential government review processes for new AI models. This engagement underscores Anthropic’s role in shaping U.S. AI policy amid capability concerns.
Julian E. Barnes: Julian E. Barnes is a New York Times reporter specializing in U.S. intelligence agencies and international security matters. He reported on the Trump administration’s internal deliberations over AI policy shifts toward government review processes. His expertise informs analysis of national security implications in technology governance.
Trump administration: The Trump administration oversees the executive branch of the United States government under President Donald Trump following his return to office in 2025. Initially favoring a hands-off approach to AI to bolster competition against China, it rolled back prior regulatory measures but is now deliberating oversight including pre-release vetting of AI models. This involves plans for an executive order establishing an AI working group with tech executives and officials.

`json
{
“UK Precedent”: “Britain utilizes multiple government bodies to conduct reviews ensuring AI models comply with safety standards before deployment.”,
“Policy Reversal”: “The Trump administration swiftly revoked Biden’s 2023 executive order on AI safety evaluations upon taking office but is now contemplating a new framework for government reviews.”
}
`