Anthropic Adviser Warns AI Biological Weapons Risk Is Real

An adviser to AI company Anthropic has publicly stated that the risk of artificial intelligence facilitating biological weapons development is "not hypothetical," marking a significant acknowledgment of concrete AI safety concerns from within the industry.

Ben Buchanan, a Johns Hopkins professor who previously advised President Biden and now serves as an adviser to Anthropic, made the statement during an appearance on CBS's "Face the Nation with Margaret Brennan." He was joined by Chris Krebs, who led the Cybersecurity and Infrastructure Security Agency (CISA) during the first Trump administration and now contributes to CBS News.

The regulation debate

The discussion centered on whether government regulation of AI is necessary and how such oversight should be structured. The conversation comes as policymakers worldwide grapple with balancing AI innovation against emerging risks, particularly in sensitive domains like biosecurity.

Buchanan's role is notable: he straddles both academic research at a major university and advisory work for Anthropic, one of the leading AI safety-focused companies. His statement suggests that concerns about AI accelerating biological threats have moved from theoretical risk assessments to concrete policy considerations.

Why it matters

When advisers to major AI companies publicly acknowledge specific catastrophic risks as real rather than speculative, it signals a shift in how the industry approaches safety disclosures. Buchanan's statement may influence regulatory discussions as lawmakers determine whether voluntary industry commitments are sufficient or if mandatory safeguards are needed for AI systems that could enable dangerous biological research.

Industry and government perspectives converge

The pairing of Buchanan with Krebs—a figure known for defending election security infrastructure—suggests CBS framed the segment around critical infrastructure protection and national security implications of AI systems. Both panelists bring experience working at the intersection of technology policy and government security concerns.

Anthropic has positioned itself as a safety-conscious AI developer, implementing measures like Constitutional AI to align systems with human values. The company's willingness to have an adviser speak publicly about biological weapons risks indicates these concerns factor into its development approach.

The details were first reported by CBS News in their June 7, 2026 broadcast of "Face the Nation with Margaret Brennan."