MSTRThe dangers of AI, documented. Real incidents, real sources, no science fiction. A running record of where AI has already caused harm.https://mstr.ai/Automation bias: when "the computer said so" overrides judgementhttps://mstr.ai/blog/automation-bias/https://mstr.ai/blog/automation-bias/The most documented AI failure mode is not the machine being wrong. It is humans trusting it anyway.Thu, 26 Mar 2026 00:00:00 GMTThe harm is already herehttps://mstr.ai/blog/the-harm-is-already-here/https://mstr.ai/blog/the-harm-is-already-here/You do not need superintelligence for AI to ruin a life. The documented damage is mundane, bureaucratic, and ongoing.Wed, 11 Mar 2026 00:00:00 GMT