Pentagon Official Dismisses Anthropic Engineer's AI Doomsday Warnings
US Department of Defense Chief Technology Officer Emil Michael publicly dismissed warnings from former Anthropic engineer Jacob Coxon regarding AI existential risks, attributing public fear of AI doom to anxiety over technological revolution. Michael argued that fears of superintelligent AI and mass job losses form a doom loop that can be effectively managed through market mechanisms, government involvement, and industry collaboration. This clash underscores the growing ideological rift between US national security leaders prioritizing rapid AI adoption and safety researchers warning of unchecked superintelligence. It also highlights escalating tensions between the Pentagon and Anthropic following failed negotiations over military safety guardrails that led to the company being designated a defense supply chain risk. Coxon warned that AI labs are racing toward recursive self-improving superintelligence capable of hacking systems and gaining real-world resources overnight. Meanwhile, Michael noted that Anthropic remains designated as a defense supply chain risk as both parties await a decision from the DC Circuit Court of Appeals regarding Anthropic's refusal to grant unrestricted military access to its models.
## BACKGROUND
AI existential risk refers to the hypothetical scenario where advanced artificial general intelligence could cause human extinction or global catastrophe if its goals become misaligned with humanity. A core vector of this concern is recursive self-improvement, a process where an AI recursively rewrites its own code to exponentially increase its intelligence beyond human oversight.