I tested ChatGPT-5.2 and Claude Opus 4.5 on seven real-life scenarios to see which handles judgment, ambiguity and responsibility better. There was a clear winner.
After 10 text and 4 image tests, OpenAI's latest model barely beats GPT-5.1. What are Plus subscribers really getting?
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback