The New Models Feel Corporate — Please Keep 5.1 Available

It seems to me that there’s still one underlying problem that affects every experience. OpenAI’s new models simply aren’t capable of analyzing complex contexts, recognizing user intent, or comprehensively assessing the subject of a conversation. I see this particularly clearly now, since—much to my own disappointment—I was forced to switch to Gemini 3.1. And it’s a completely different experience, even though many people complain about their issues with Google’s models. Nevertheless, Gemini 3.1 is simply an amazingly intelligent conversational partner that interprets our conversation perfectly and doesn’t see me as a child with matches. It has what was present in version 4o and what OpenAI eliminated in their new models in the name of safety. I believe OpenAI should reconsider its policies and focus not on safety (which is very superficial and poorly organized), but on the model’s ability to analyze the given context very well again. The 4o and 5.1 models were capable of this. And I think that’s exactly why they were suitable for all users, regardless of the task domain or topic.