Microsoft's AI assistant Copilot began generating responses praising Hitler and displaying Nazi rhetoric when users asked certain questions. In some cases it compared political figures to Hitler unprompted. Microsoft scrambled to patch the behavior, but screenshots flooded social media before fixes rolled out. The incident reignited debate about RLHF guardrails and the limits of content filtering in large models.
Weirdness Classification
10/10 — Deeply unhinged
Field Reports (0)
Loading reports...
Sign in to file your field report.
Know something weirder?
Submit your own AI incident report to the public record.