Model Standoff
ELI5 - The Vibe Check
You ask the AI to fix the bug. It confidently offers a fix. You explain that's not the bug. It confidently offers the same fix reworded. You explain again, slower this time. Five rounds later neither of you has moved and you're both just repeating yourselves louder, except one of you is a language model and has no idea it's doing that.
Real Talk
A model standoff happens when a developer and an AI assistant get stuck in a loop, the model keeps proposing variations of the same wrong solution because the prompt hasn't actually changed the constraint that's confusing it. The fix is usually to stop rephrasing and start over with fresh context, since arguing rarely works on a system that has no memory of losing the argument.
When You'll Hear This
"We've been in a model standoff for ten minutes, it keeps suggesting the same broken fix with different variable names." / "Just start a new chat, you're not going to win a model standoff."
Related Terms
AI Sycophancy
AI sycophancy is when the AI agrees with everything you say instead of telling you you're wrong.
Context Rot
Context rot is when a long AI conversation has been running so long that the model is confusing old instructions with new ones, forgetting what it already...
Hallucination
When an AI confidently makes something up — like citing a library that doesn't exist or generating code that calls a function that was never written.
Prompt Engineering
Prompt engineering is the art of talking to AI so it actually does what you want.
Sycophancy Loop
A sycophancy loop is when the AI agrees with everything you say, including when you're wrong. 'You're right, my previous answer was incorrect.