OpenAI's Sam Altman Defends Pentagon AI Deal Amid Employee Pushback

Top Summary

• What happened: OpenAI CEO Sam Altman defended the company's Pentagon deal for classified AI deployment, stating employees don't get a vote on US military operations.
• Why it matters: The deal sparked internal and external criticism regarding timing and ethical implications, especially after rival Anthropic's blacklisting.
• What changes for people: OpenAI engineers will be embedded with government teams, but operational decisions remain with the Pentagon.
• Who is affected: OpenAI employees, the AI safety community, the US military, Anthropic, and potentially NATO.

Pentagon Deal Sparks Controversy

OpenAI CEO Sam Altman addressed employee concerns regarding the company's recent deal with the Pentagon to deploy AI models on classified networks.

The deal, announced shortly after rival Anthropic was blacklisted by the Department of Defense, drew immediate backlash.

Altman emphasized that employees have no say in US military operations, regardless of their personal views. He stated,

"So maybe you think the Iran strike was good and the Venezuela invasion was bad. You don't get to weigh in on that."

 

Timing and Optics Questioned

The timing of the announcement, late on a Friday evening, and its proximity to Anthropic's blacklisting, raised eyebrows.

Altman admitted the announcement looked "opportunistic and sloppy." He acknowledged that the issues are complex and demanded clearer communication.

Some OpenAI employees publicly criticized the move, with many having recently signed an open letter supporting Anthropic's red lines.

Altman Defends OpenAI's Position

Despite the criticism, Altman defended OpenAI's involvement, stating the Pentagon respects the company's safety measures.

He clarified that while OpenAI engineers will be embedded with government teams, the Pentagon retains operational control.

Secretary Pete Hegseth, not Sam Altman, makes day-to-day military decisions.

Competition and Future Plans

Altman acknowledged the competitive landscape, suggesting some competitors, like xAI, might offer fewer restrictions.

OpenAI is now exploring a contract to deploy on all NATO classified networks, solidifying its position in the defense sector.

While Anthropic's Claude was reportedly used in recent military operations, Altman reiterated that Anthropic should not be labelled a supply chain risk.

What to Watch Next

It remains to be seen whether Anthropic will challenge its blacklisting, and if other AI companies will receive similar deal terms as OpenAI. The development of AI in the defense sector and the ethical implications of its use will continue to be closely monitored.