Claude Is a Contrarian
A user reports that the AI model Claude consistently ignores specific instructions to avoid contradictions and unnecessary additions in its output. The author notes that this 'contrarian' behavior makes Claude's generated content easily identifiable compared to other LLMs.
Why it matters
This highlights the challenges of 'steerability' and instruction following in advanced AI models, which is a critical concern for developers and power users.
LLM Claude Content Creation Programming Claude is a Contrarian RDX 3 min read · 1 hour ago --
Of all the LLMs I’ve used and driven daily, Claude is the one that absolutely contradicts my instructions.
No, it’s not about the inconsistent routing, nerfing, or asking for a model and not knowing what you got. This is not even about quality of code. LLMs have matured to the point where, even underpowered models are dependable. Deepseek Flash and OpenAI Luna would show that for you.
I can tell it in my CLAUDE.md: “Don’t contradict sentences”. I can be damned sure it will sprinkle contradictions everywhere:
This wonderful feature does this , not that . Feature that move you , not bore you. Feature that gives you hope , never disillusion. This was supposed to be the problem. It’s not.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in