OpenAI recently entered into a collaboration with the U.S. Department of Defense, according to official sources. The collaboration enables the use of the company's AI technology in the U.S. and Israel's war against Iran, the company said. Attacks by the U.S. and Israel against Iran have been assessed by several experts as violations of international law. Criticism that OpenAI did not seem as restrictive led to CEO Sam Altman changing the agreement with points that the technology will only be used within legal frameworks and not have the intention to surveil U.S. residents, official sources reported. The specific legal frameworks or restrictions in the revised agreement remain unclear.
In a separate partnership, OpenAI began a collaboration with Schibsted, the Norwegian media conglomerate that in Sweden includes Svenska dagbladet, Omni, and Aftonbladet, in February last year, according to official sources. The collaboration involves OpenAI training its material on Schibsted media's content for two years, and Schibsted gets access to the tech company's insights and technology in return. The tech company pays a small part of Schibsted journalists' salaries, official sources said. Schibsted is not in a dependent position on OpenAI, according to Fredric Karén, Group Director at Schibsted Sweden. The partnership agreement clearly states how the conglomerate's content may be used, Karén described.
Another AI company, Anthropic, withdrew from collaboration with the U.S. government after the Department of Defense requested a deeper partnership, official sources reported. Anthropic's CEO raised concerns that their technology would be used for mass surveillance and implemented in autonomous weapons systems. Many companies have collaborations with American companies that supply the U.S. military with components, according to Fredric Karén, Group Director at Schibsted Sweden.