The big names in the artificial intelligence industry don't always get along. We've seen lawsuits between AI companies and intense rivalries between leaders that turn into all-out feuds. But it seems that they generally agree on at least one thing: AI should not be used to create biological weapons.

CEOs of some of the world's leading AI companies signed a public letter this week urging governments to address the risks that could come from bad actors using their technology. The letter encourages Congress to enact laws to improve the tracking of synthetic DNA sequences that could be used to create biological weapons. 

Signed by OpenAI's Sam Altman, Anthropic's Dario Amodei, Meta's Alexandr Wang, Microsoft AI's Mustafa Suleyman, Google DeepMind's Demis Hassabis and other scientists and AI lab leaders, the letter suggested legislation to require companies that sell synthetic DNA and manufacturers of synthesis machines to thoroughly check "sequences of concern and to verify customer legitimacy before shipping orders."

The letter, also signed by leaders in the synthetic DNA industry and experts in national security, makes it clear that AI accelerates the threat of biosecurity threats.

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

"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 has a distinct sense of urgency, requesting that lawmakers implement guardrails swiftly. "This is a rare moment of agreement across stakeholders that are often at odds," it reads. "We hope policymakers will meet it with decisive action."

Why synthetic DNA is a real threat 

Synthetic DNA -- short strands of nucleic acid produced by commercial DNA synthesizers -- was first developed in a test tube by biochemist Arthur Kornberg in 1958, winning him a Nobel Prize in the process. Synthetic DNA has been used to understand the basis of life, to create biological tools and medical treatments, like vaccines and gene therapies and to discover how genetic code translates into proteins. Synthetic DNA has led to massive breakthroughs in medicine, agriculture and science since the 1950s. 

Since then, it has become far cheaper and faster to make. It's this decline in cost plus AI automation that is changing who can design and build biological agents. Ordered as oligonucleotides -- short strands of DNA or RNA -- or larger gene fragments, synthetic sequences can be assembled into viruses or used to alter organisms' traits. While there are many benefits to having synthetic DNA used in research, the biggest danger is that bad actors could engineer deadly biological weapons or "superviruses." 

Most commercial providers already screen many orders for known dangerous sequences, but benchtop DNA synthesizers and automated tools make it easier for non-experts to obtain and assemble genetic material than ever before.

The increased risk today stems from two main sources: More sophisticated AI systems that can design, optimize, troubleshoot or simulate biological systems and increasingly accessible DNA synthesis and lab automation. In the worst case, powerful models could shorten the expertise and trial-and-error required to create or modify pathogens. Or they could help craft novel biological weapons with higher transmissibility and virulence. 

Even if highly sophisticated, supervirus-caused pandemics remain unlikely, lower-barrier misuse -- such as recreating eradicated agents, modifying viruses or producing novel, dangerous organisms -- is feasible and alarming. 

What are the current safety measures? 

Current guardrails are a mix of industry best practices, voluntary screening and government regulation. Many synthesis companies willingly participate in sequence screening initiatives that flag orders matching known pathogens or dangerous sequences. Some generative AI companies also apply content and model-safety policies that block prompts about biological-harm instructions.

But in the open letter, experts across the fields of science, technology, public policy, academia and law are proposing tougher measures. 

The public letter shows a consensus that the risks of AI-developed bioweapons deserve urgent attention and stronger governance before capabilities outpace current policy -- which is often the case with generative AI. 

Public appeals like the letter can pressure policymakers, funders and company executives to accelerate practical mitigations and guardrails, including harder synthesis access, model audits, monitoring and incident reporting. With these precautions and safety measures in place, experts aim to reduce the odds that advancements in AI will enable accidental or malicious biological events by the time more permanent regulations are in place.