Palantir CTO Says Human Error Likely Behind Deadly Iran Strike

what happened

Benzinga reports that Palantir Technologies Inc. CTO Shyam Sankar rejected claims that the company’s AI software contributed to the Feb. 28 strike on a school in Minab, Iran. At least 123 schoolchildren were killed. Speaking on CBS’s Face the Nation on Sunday, Sankar said AI can accelerate military planning, including operations involving 1,000 strikes, while helping personnel check no-strike lists and identify sites near civilian infrastructure.

Sankar described as “inaccurate” a report alleging that military personnel depended too heavily on Palantir’s Maven system during attacks in Iran, and called for the final report. He said “the likely cause is human error,” without presenting a definitive conclusion. Palantir also quickly built additional AI agents to conduct independent checks and lower the risk of another incident.

A March report said a U.S. military investigation was examining an incident involving a Tomahawk missile that hit Minab’s Shajarah Tayyebeh elementary school on Feb. 28. That account pointed to outdated data and basic human error rather than AI. The discussion comes as armed forces expand their use of autonomous and AI-enabled systems. Sarah Yager, OpenAI’s first human rights and responsible deployment lead, said “everyone is very worried” about AI in armed conflict and autonomous weapons. She argued that human rights and international law should guide military AI use, rather than standards set solely by technology companies.

why it matters

The episode puts additional focus on accountability, data quality and human oversight in military AI. For markets, it highlights reputational and regulatory risks for Palantir and other defense-technology providers.

instruments