The big test for Anthropic
Anthropic faces one of the most critical moments in its history, between an unprecedented confrontation with the US Government and internal security failures that expose the company’s weaknesses.
Founded by former OpenAI employees in 2021, Anthropic is going through one of the most turbulent periods in its short history. Punished by the United States Government, the maker of the popular chatbot Claude decided to go to court to assert its rights. And in recent days it has been able to claim a ‘semi-victory’, after winning an injunction against the US Pentagon.
The conflict arose from Anthropic’s refusal to allow its artificial intelligence (AI) models to be used in autonomous weapons systems and in the mass surveillance of American citizens. As a result, the company lost a 200 million dollar contract with the Department of War (formerly the Department of Defense) and was put on a Pentagon ‘blacklist’ that bars certain public bodies from using its services.
AI translation from the Portuguese. First published in ECO on 2 April 2026.