Anthropic CEO Dario Amodei calls OpenAI’s messaging around military deal ‘straight up lies,’ report says

Anthropic CEO Dario Amodei has launched a blistering public attack on rival OpenAI, accusing CEO Sam Altman of engaging in 'safety theater' and spreading 'straight up lies' regarding OpenAI's new contract with the U.S. Department of Defense (DoD). The feud erupted after Anthropic, which already had a $200 million military contract, failed to reach a new agreement with the DoD because it insisted on explicit prohibitions against using its AI for domestic mass surveillance or autonomous weaponry. The DoD instead struck a deal with OpenAI, which Altman claimed included similar protective red lines. Amodei's internal memo, reported by The Information, sharply criticizes this narrative, accusing Altman of falsely 'presenting himself as a peacemaker and dealmaker' while prioritizing 'placating employees' over genuine safety.

The core dispute centers on contract language: the DoD insisted on access for 'any lawful use,' a clause Anthropic rejected but OpenAI accepted. OpenAI stated it secured explicit confirmation that 'lawful use' excludes mass surveillance, but critics argue laws can change. The public reaction has been significant, with reported 295% spikes in ChatGPT uninstalls and Anthropic's app climbing to #2 in the App Store. Amodei noted the 'attempted spin' is failing with the public and media but expressed concern it might influence OpenAI's own staff. This clash highlights the deepening ethical rift in AI governance between competing corporate philosophies as military adoption accelerates.