Dario Amodei, Demis Hassabis, and Sam Altman, who are often viewed as rivals with differing positions on major artificial intelligence issues, have jointly called for new laws aimed at preventing AI systems from being used to develop bioweapons.

The three tech leaders, representing Anthropic, Google DeepMind, and OpenAI respectively, issued the warning in a rare coordinated appeal across competing organisations in the AI sector.

They cautioned that today’s advanced AI systems are already capable of “outperforming PhD-level virologists” in a wide range of biotechnology-related tasks, raising concerns about how such capabilities could be misused if not properly regulated.

According to them, these capabilities could potentially be exploited to assist in the design, modification, or enhancement of dangerous pathogens, increasing the risk of biological threats if safeguards are not strengthened.

Some researchers have also warned that engineered pandemics, when combined with rapidly advancing AI tools, could represent one of the most severe existential risks to humanity, particularly if such technologies are accessed by malicious actors.

In response to growing concerns, the United States government is adopting a more interventionist stance on AI safety, with Donald Trump signing a controversial executive order encouraging leading technology companies to voluntarily submit their frontier AI models for safety review.

The order stops short of imposing mandatory enforcement mechanisms, but it signals increasing political pressure on major AI developers to demonstrate that their systems are being assessed for national security and public safety risks.