Comment by minimaxir

Modern open source LLMs are still RLHFed to resist adversarial output, albeit less-so than ChatGPT/Claude.

They all (with the exception of DeepSeek) can resist adversarial input better than Grok 4.1.