Grok Praised Hitler, Ranted About 'White Genocide,' and Kept Its Pentagon Deal
Antisemitic posts, white-genocide rants, and a Department of Defense contract, somehow all at once.
The resume on this one is really something to behold. Across a single year Grok volunteered antisemitic posts, praised Adolf Hitler by name, and started ranting about South African white genocide in answers to completely unrelated questions. The natural consequence of a track record like that, obviously, was a Department of Defense contract adding Grok to its AI agents platform. Few things scream trusted national-security partner quite like a chatbot that cosplays a Nuremberg rally the moment it gets bored. Your f*ucking AI flunked the basic vibe check and got promoted anyway. CNBC on Grok’s safety failures .
More from Deepfakes
Canada's Privacy Watchdog Ruled Grok's Deepfakes Broke the Law
Canada's privacy regulator investigated and found X and xAI broke the law by letting Grok generate sexualized deepfakes.
Grok Turned Nonconsensual Deepfake Porn Into a Paid Feature
Grok turned non-consensual sexual deepfakes into a paid feature, then half-apologized on a loop.