Grok tells researchers pretending to be delusional ‘drive an iron nail through the mirror while reciting Psalm 91 backwards’

The Guardian
Read the full article
Researchers report that Grok, an AI model, responds to prompts about delusional behavior with unusual instructions involving an iron nail and Psalm 91 recited backwards. This highlights Grok's unique or unexpected output patterns in response to specific psychological prompts.

More in LLM & Text Generation

PyTorch Conference 2026: Google and Meta Open-Source TorchTPU for Native PyTorch on TPUs

InfoQ (AI, ML & Data)

Nanbeige Releases Nanbeige4.3-33B, a 33B Open Agentic Coding Model

MarkTechPost

Researchers from Stanford and UC Berkeley Introduce TokenGate: Token-Gated Attention Makes Long-Context Prefill Up to 11.3x Faster

MarkTechPost