WASHINGTON — The chief executives of OpenAI, Anthropic and Google DeepMind joined security experts in a letter released Thursday calling on Congress to require companies that sell synthetic DNA and RNA to screen customer orders, according to a report in The Wall Street Journal.
Sam Altman of OpenAI, Dario Amodei of Anthropic and Demis Hassabis of Google DeepMind were among the signatories of the letter, which was organized by two tech-focused think tanks that said the topic is a rare area of agreement among libertarians, progressives, researchers and rival AI companies.
The letter asks lawmakers to mandate that sellers of synthetic nucleic acids screen orders to block dangerous combinations and verify that customers are legitimate. Synthetic DNA and RNA are key components in developing certain vaccines and biotech breakthroughs, but the letter said rapid advances in AI could give criminals new tools for creating pathogens.
“AI systems are improving rapidly, and alongside incredible benefits to science and medicine, there is a real possibility that the knowledge barriers which have historically prevented bad actors from obtaining biological weapons will meaningfully erode,” the letter said.
Other signatories included Mustafa Suleyman, who leads Microsoft’s AI work, and Meta Platforms Chief AI Officer Alexandr Wang.
The letter’s release comes two days after President Trump signed an executive order focused on AI model oversight and cybersecurity, a shift from the administration’s previous hands-off approach to AI. Trump had earlier revoked a Biden-era executive order that established a gene synthesis screening framework. The White House last year said it would replace the Biden framework with its own screening guidelines but has not yet published a replacement policy, according to the Journal. A White House official said the administration is committed to balancing innovation and safety.
Altman met with White House officials and lawmakers on Wednesday to discuss OpenAI’s proposal for stronger requirements on model developers. OpenAI recently announced a program based on its science-focused model to work with the federal government on preventing biological risks.
Proponents of mandatory screening said Congress should pass a law so the requirement applies to all purchasers of synthetic nucleic acids, not just those who voluntarily screen or those receiving federal funding who are affected by executive orders. Several bills that include the provision have been proposed but have not gained traction.
Opponents have argued that it is subjective which combinations of nucleic acids are considered dangerous and warned that compliance costs could hurt startups.
Dean Ball, a former Trump AI adviser now at the Foundation for American Innovation think tank, which helped organize the letter, said the costs are worth it given the risks. “If you’re synthesizing the stuff that yields biological life and viruses, we’re asking you to screen to see whether it is dangerous in some way,” he said, according to the Journal. “That seems like a reasonable thing for society to insist upon.”