Pentagon investigators found that overreliance on Palantir's Maven AI system, combined with flawed intelligence and outdated imagery, contributed to a February missile strike in Iran that killed 123 children.
US officials revealed the findings in a review of the operation, identifying multiple failures in the targeting process. The Maven system, an AI platform used for intelligence analysis, was among several factors that led to the strike.
Investigators cited flawed intelligence assessments and outdated satellite imagery as additional contributors to the incident. The combination of these issues—particularly heavy reliance on the AI system without sufficient human verification—created a critical gap in the decision-making process.
The incident underscores growing concerns about AI deployment in military operations. Defense officials have increasingly adopted automated systems to process vast amounts of intelligence data, but the findings suggest such tools require robust oversight and independent verification.
Palantir Technologies did not immediately respond to requests for comment. The Pentagon's review recommends strengthening human oversight protocols and improving intelligence verification procedures before relying on AI-generated assessments for high-stakes military decisions.
Virginia Gov. Abigail Spanberger signed an executive order this week that will slow approvals for new data center projects and establish a task force to study AI's impact on the workforce.
According to CNN sources, artificial intelligence almost caused the US military to attack a Chinese vessel it incorrectly identified as carrying nuclear weapon components. The incident highlights growing risks as militaries increasingly rely on AI systems for critical decisions.
World model companies are raising significant capital and generating substantial buzz, yet founders and data suppliers remain largely tight-lipped about their actual progress and capabilities.
Anthropic is partnering with Accenture to embed evaluators directly within its operations to test the safety of its advanced AI models. The move comes as the AI safety startup projects annualized revenue exceeding $100 billion in 2026.