Top scientists at OpenAI and Anthropic ask U.S. for tools to pace AI development - NBC News

What the report says

NBC News reported that more than 1,000 employees at major artificial intelligence companies signed a statement urging the U.S. government to support international work on ways to “deliberately pace” the development of frontier AI systems. The signatories included senior figures and co-founders from OpenAI, Anthropic and Google DeepMind, according to NBC, as well as chief scientists or science officers Jared Kaplan of Anthropic, Jakub Pachocki of OpenAI and Shengjia Zhao of Meta. Bloomberg had earlier reported that the letter was circulating among AI company workers.

The statement asks for technical and governance mechanisms that could give governments, companies and society more time to manage emerging risks, improve security and build stronger oversight. NBC reported that the letter does not directly demand an immediate slowdown or pause in AI development; rather, it calls for ensuring that such a tool could be available if needed as automated AI research advances.

The appeal comes amid rising concern that AI systems are beginning to assist with, or partially automate, tasks involved in building more capable models. NBC cited recent claims from leading AI companies about AI’s role in developing future systems, while also noting experts’ distinction between AI-assisted research and genuine independent self-improvement. Anthropic has previously said society may need the option to slow or temporarily pause frontier development if automated AI research becomes more powerful.

The issue matters because frontier AI is increasingly tied to national security, cybersecurity and international competition. NBC reported that the letter calls for U.S. involvement in global coordination on policy and technical safeguards, even as the Trump administration has often opposed international AI governance efforts on the grounds that they could limit American innovation.