Pentagon Summons Anthropic CEO Dario Amodei Amid Elon Musk’s Sharp Criticism

According to reports, US Defence Secretary Pete Hegseth has called a "high-stakes" and "tense" meeting at the Pentagon with Dario Amodei, CEO of AI firm Anthropic, over the military's utilisation of the Claude AI model. Reportedly, Elon Musk criticised the AI firm for allegedly taking data used for training.

Although Hegseth pushed AI companies to ease constraints on their technology, Anthropic refused, leading media sources to suggest that the meeting was "not a friendly" one. Following the signing of a $200 million pilot contract last year, Claude became the sole AI system to be integrated into classified defence networks.

Anthropic Refused US Defence Sec Request

To free their AI from contractual constraints, Hegseth requested in a January 9 memo that the companies involved rewrite the agreements. On the other hand, Anthropic has refused to budge from its stance on the issue of completely autonomous weaponry and limitations on widespread surveillance of Americans. The military has issued a warning that Anthropic might be labelled a "supply chain risk", which would render contracts null and void. So, even though the AI firm's representative hailed the talks as "productive", other Pentagon partners are still barred from utilising Claude.

According to the reports, Anthropic's extensive integration into defence systems makes its replacement a complicated task. According to the New York Times, the Pentagon is putting pressure on Anthropic by signing agreements with xAI, Elon Musk's artificial intelligence company. Further, the Pentagon has almost finalised a contract with Google for the Gemini model. Anthropic claimed three Chinese AI companies stole millions of Claude outputs through chatbots and used them to train their own models before the conference.

Musk not Happy with Anthropic

Disputes broke out among the AI participants when Anthropic made certain assertions. Musk retaliated by criticising Anthropic on X, stating, "Anthropic is guilty of stealing training data at massive scale and has had to pay multi‑billion-dollar settlements for their theft."

Anthropic's next-gen AI assistant Claude is developed and educated to be secure, accurate, and safe. Reviews of legal documents, compliance checks, sales planning, analyses of marketing campaigns, reconciliation of finances, visualisation of data, reporting based on SQL, and enterprise-wide document search might all be automated by the new AI assistant. Not only did Elon Musk, but also netizens, criticise Anthropic for its supposed hypocrisy. Anthropic was accused by netizens of using the free internet to train their Charles AI model.