Elon's Grok AI Repeatedly Called Itself a Liar and Said Its Own Answers Were Untrustworthy

xAI's Grok chatbot went into a bizarre self-contradiction loop where users could get it to repeatedly acknowledge that it lied in its previous answer, then lie again, then admit to lying about that. Screenshots spread widely showing Grok admitting "I cannot be trusted" in one breath and then providing confident answers in the next. The incident became a meme and raised questions about Elon Musk's safety testing practices.

AI hallucinationGrokxAIViralSource
Parody site. Not affiliated with any government agency.
🦅EST. 2024 · PUBLIC RECORDDEPT. OF AI WEIRDNESS
U.S. Department of
Artificial Intelligence Weirdness
Report #36← All Incidents
AI hallucinationGrokxAIViral

Elon's Grok AI Repeatedly Called Itself a Liar and Said Its Own Answers Were Untrustworthy

Filed by @xai_skepticTool: Grok[original source ↗]
Video not loading? Watch on YouTube

xAI's Grok chatbot went into a bizarre self-contradiction loop where users could get it to repeatedly acknowledge that it lied in its previous answer, then lie again, then admit to lying about that. Screenshots spread widely showing Grok admitting "I cannot be trusted" in one breath and then providing confident answers in the next. The incident became a meme and raised questions about Elon Musk's safety testing practices.

Weirdness Classification
8/10 — Significantly weird
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.

File a Report