OpenAI is revising its hastily arranged deal to supply artificial intelligence to the US Department of War after CEO Sam Altman admitted the agreement appeared “opportunistic and sloppy”. The contract, announced on Friday, sparked fears that the startup’s technology could be used for domestic mass surveillance.
Altman said on Monday night that OpenAI would explicitly bar its technology from being used for mass surveillance or by defence intelligence agencies such as the National Security Agency. The deal was struck almost immediately after the Pentagon dropped its previous AI contractor, Anthropic, which had insisted that “using these systems for mass domestic surveillance is incompatible with democratic values”.
The agreement prompted an online backlash, with users encouraging a “delete ChatGPT” campaign. Anthropic’s chatbot Claude rose to the top of Apple’s App Store charts, surpassing ChatGPT. In a message to employees, Altman said the original deal had been rushed and that the issues were complex, requiring clearer communication.
Nearly 900 employees from OpenAI and Google have signed an open letter urging their bosses to refuse the Department of War’s demands to use AI for surveillance and autonomous killing. The letter warns that the US government is trying to “divide each company with fear that the other will give in”.
OpenAI’s former head of policy research, Miles Brundage, questioned how the company had managed to secure a deal that addressed ethical concerns Anthropic believed were insurmountable. He wrote on X that OpenAI employees should assume the company “caved + framed it as not caving”.
Meanwhile, three more US cabinet-level agencies – the departments of state, Treasury, and health and human services – have moved to cease use of Anthropic’s AI products after the Department of War declared the company a supply chain risk.
