{"slug": "ask-hn-is-codex-with-gpt-5-5-extra-high-being-dumbed-down", "title": "Ask HN: Is Codex with GPT 5.5 Extra High being dumbed down?", "summary": "A user on Hacker News reports that OpenAI's Codex with GPT-5.5 Extra High has degraded in quality, comparing the shift to replacing a reliable tech lead with an overconfident junior developer. The user claims the model no longer thinks through problems or follows instructions, and that the change is binary and cannot be fixed with better prompting.", "body_md": "Hi HN, just want to rant and see if anybody can relate.\n\nThe product is not the same as i signed up for few month ago and the same shift i've experienced with Claude Code on Opus 4.6-4.7\n\nThe best way to describe the difference is you hire a reliable 'intelligent' tech lead who 'gives a shit' and in some time you eventually get over-confident junior dev that acts as destructive token burner.\n\nNo thinking during the process, not even really following instructions.\n\nIt's pretty much a binary shift that happens and in my experience can't be cured with prompting.\n\nThere is no way to actually tell what's happening under the hood with models but the difference is noticeable from the first reply.\n\nI'm on max, GPT 5.5 extra high, always mindful of context window, use plan mode etc.\n\nIs it the model: dumbing down or quietly route to another model for some reason? Is it the harness? or is it my imagination?\n\nCurious what's your recent experience.\n\nPS: frontier models should add \"give-a-shit: max\" along with 'thinking: max', 'effort: max' and make them actually work.\n\nComments URL: [https://news.ycombinator.com/item?id=48734115](https://news.ycombinator.com/item?id=48734115)\n\nPoints: 2\n\n# Comments: 0", "url": "https://wpnews.pro/news/ask-hn-is-codex-with-gpt-5-5-extra-high-being-dumbed-down", "canonical_source": "https://news.ycombinator.com/item?id=48734115", "published_at": "2026-06-30 15:27:25+00:00", "updated_at": "2026-06-30 15:50:13.464522+00:00", "lang": "en", "topics": ["large-language-models", "ai-products", "ai-tools"], "entities": ["OpenAI", "Codex", "GPT-5.5", "Claude Code", "Opus 4.6", "Opus 4.7", "Hacker News"], "alternates": {"html": "https://wpnews.pro/news/ask-hn-is-codex-with-gpt-5-5-extra-high-being-dumbed-down", "markdown": "https://wpnews.pro/news/ask-hn-is-codex-with-gpt-5-5-extra-high-being-dumbed-down.md", "text": "https://wpnews.pro/news/ask-hn-is-codex-with-gpt-5-5-extra-high-being-dumbed-down.txt", "jsonld": "https://wpnews.pro/news/ask-hn-is-codex-with-gpt-5-5-extra-high-being-dumbed-down.jsonld"}}