Sam Altman and three other major AI leaders have signed a public letter calling for new laws to prevent artificial intelligence from being used to develop biological weapons. The move represents an unusual moment of unity among competitors who typically clash over strategy and safety priorities. Altman was joined by Anthropic CEO Dario Amodei, Google DeepMind's Demis Hassabis, and Microsoft AI's Mustafa Suleyman in supporting mandatory screening requirements for companies selling synthetic DNA.

Why Are AI Leaders Concerned About Bioweapons Now?

The letter, titled "In support of mandatory nucleic acid synthesis screening and recordkeeping," explains that artificial intelligence systems have reached a concerning capability threshold. According to the signatories, AI now outperforms PhD-level virologists on questions about highly technical laboratory procedures within their own domains of expertise. This represents a meaningful shift in the knowledge barriers that have historically protected against biological weapons development.

"While the issue is not new, the pace of progress in artificial intelligence is. AI systems now outperform PhD-level virologists on questions about highly technical laboratory procedures in their own domains of expertise. 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 states.

Sam Altman, Dario Amodei, Demis Hassabis, and Mustafa Suleyman, in public letter

The concern centers on how AI could lower the technical bar for creating dangerous pathogens. If synthetic DNA suppliers lack screening mechanisms, bad actors could potentially misuse genetic material to create harmful bioweapons. The letter describes this moment as "a rare moment of agreement across stakeholders that are often at odds".

What Makes This Agreement Surprising?

The unity is particularly notable given recent tensions between Altman and Amodei. The two leaders have clashed over AI safety and military applications. Most recently, Amodei stated he could not "morally capitulate" to demands from the US Department of War to use Anthropic's technology for surveillance and autonomous weapons. OpenAI subsequently struck a deal with the Pentagon, though Altman later admitted in an internal memo that the announcement looked "opportunistic and sloppy".

The personal friction between the two CEOs was visually evident at the India AI Impact Summit in New Delhi earlier this year, where they were the only two leaders in a 14-person group photo who did not hold hands. Amodei previously worked at OpenAI before leaving in 2021 due to concerns about AI safety, and he founded Anthropic that same year alongside six other former OpenAI employees.

How Can Companies Prevent AI-Enabled Bioweapon Development?

• Synthetic DNA Screening: Companies selling synthetic DNA should implement mandatory screening of customers and orders to identify potentially dangerous requests before fulfilling them.
• Recordkeeping Requirements: Maintaining detailed records of synthetic DNA sales and customer information would create accountability and help authorities track suspicious patterns.
• Industry Standards: Establishing shared protocols across synthetic DNA suppliers would prevent bad actors from simply moving to less cautious vendors.

The letter's focus on synthetic DNA screening is strategic because it targets a critical chokepoint in bioweapon development. Rather than trying to control AI itself, the signatories are proposing regulations on the physical materials that AI knowledge could help weaponize. This approach acknowledges that AI capabilities cannot be easily restricted, but the supply chains for dangerous biological materials can be monitored.

The agreement among Altman, Amodei, Hassabis, and Suleyman suggests that despite their competitive differences and disagreements over military applications, AI leaders recognize a shared interest in preventing the most catastrophic misuses of their technology. The letter represents a pragmatic step toward international governance of AI-related biosecurity risks, even as the broader debate over AI safety and military use continues to divide the industry.