The Pentagon Wanted AI Weapons, But OpenAI Just Said 'No' to Their Most Controversial Request

OpenAI was about to help the U.S. military in ways that made a lot of people very uncomfortable. Now they're backing down.

The company announced it's changing its deal with the Pentagon after a wave of criticism. The original agreement would have let the military use OpenAI's technology in ways that crossed lines even for the defense industry.

Here's the deal: AI companies are in a weird spot right now. The government is their biggest potential customer, but helping build weapons makes regular people very nervous. OpenAI tried to have it both ways and got caught.

The backlash was swift. Employees complained. Customers threatened to leave. Headlines were brutal.

So OpenAI is scaling back. They'll still work with the military, but with new limits on what their AI can be used for.

This isn't just an OpenAI problem. Anthropic, the company behind Claude, is having the same fight. Their CEO publicly said they cannot agree to what the Pentagon wants.

The question everyone's asking: can AI companies serve both Silicon Valley idealists and Pentagon generals? So far, the answer seems to be no.

As reported by CBS News.

Source: CBS News