In August 1939, Albert Einstein and Leo Szilard penned a letter to President Roosevelt, warning him that Nazi Germany could be working on a nuclear weapon. They urged the U.S. to pursue research into nuclear fission. As a result, Roosevelt established a committee, which later blossomed into the Manhattan Project. So, one letter had a massive impact.
Also read: Meta Business Agent: Is your data private, who does it work for?
Fast forward to now, and we’ve got another significant letter. Sam Altman from OpenAI, Dario Amodei from Anthropic, Demis Hassabis from Google DeepMind, and Mustafa Suleyman from Microsoft AI wrote to Congress. They ask for regulations on synthetic DNA and RNA companies. These firms would need to screen clients to stop the sale of materials that could create bioweapons. While their intentions appear safeguarding, it’s still eerie thinking about how easy it could lead to something enormous and potentially harmful, just like that initial letter from Einstein and Szilard did.
The letter from the Institute for Progress and the Foundation for American Innovation lays out something that these firms have realized: Artificial intelligence is knocking down the old barriers to entry for creating bioweapons. Once upon a time, making bioweapons required specialists in microbiology, virology, and synthetic biology. Nowadays, though? Just ask an AI, and it’ll show you how to do it.
Take David Relman, a microbiologist at Stanford and one of the letter’s signatories; he cuts right to the chase. According to him, today’s AI systems can walk users through purchasing gene sequences undetected and sidestepping security checks.
Also read: At Build 2026, Mustafa Suleyman finally revealed Microsoft’s AI moves
Now, lots of companies globally earn a living by selling genetic sequences for various uses, like research and diagnostics. Although some perform their own safety checks, there’s no universal standard. A case in point is when scientists reconstructed a defunct virus back in 2017 with less than $100,000 worth of DNA bought online. As costs continue to drop, this sort of technology is getting easier for anyone to access.
That’s why the letter pushes for tough federal regulations, not just recommendations, but actual laws that everyone follows strictly. It’s a reasonable request. The hitch is in how Congress takes this info. If you show lawmakers that AI can help design dangerous pathogens, you’re not only ringing the alarm bells; you’re handing strategic data to the defense and intelligence teams. They’ll view this info as a tool for developing, investigating, and potentially exploiting it.
Scientists today face a situation somewhat like Einstein’s post-World War II era, after he had helped develop the atomic bomb. Now, these researchers aren’t mere observers; they control the tech too. This adds credibility to their warning, but it doesn’t make handling the issue any simpler. We need to block harmful AI uses, particularly with synthetic DNA. Yet, the big worry is whether rules can keep pace with tech advancements. History says probably not.
Also read: What is Microsoft’s Project Solara: The platform that wants to end the app era