{"slug": "from-gpt-5-can-t-do-basic-math-to-today-what-a-year-tells-us", "title": "From 'GPT-5 Can't Do Basic Math' to Today: What a Year Tells Us", "summary": "A year after GPT-5's release, its mathematical and reasoning capabilities have improved dramatically, with users reporting it can now walk through proofs, though some note real-world reliability issues persist on simple logic tasks.", "body_md": "# From 'GPT-5 Can't Do Basic Math' to Today: What a Year Tells Us\n\nA year ago\n\n[Next Figure AI's F.03 Ladder Climb: Autonomy or OSHA Violation? →](/en/news/4729/)\n\n## All Replies （3）\n\nC\n\nA year ago it couldn't add fractions reliably. Now it walks through proofs for me. Night and day.\n\n0\n\nZ\n\nBenchmarks improved, sure, but real-world reliability still stumbles on simple logic. Progress ≠ parity.\n\n0\n\nR\n\nHow are you catching those logic stumbles in your own testing? Curious what eval setup you use.\n\n0", "url": "https://wpnews.pro/news/from-gpt-5-can-t-do-basic-math-to-today-what-a-year-tells-us", "canonical_source": "https://promptcube3.com/en/news/4731/", "published_at": "2026-08-02 06:28:25+00:00", "updated_at": "2026-08-02 06:54:10.457844+00:00", "lang": "en", "topics": ["large-language-models", "artificial-intelligence"], "entities": ["GPT-5"], "alternates": {"html": "https://wpnews.pro/news/from-gpt-5-can-t-do-basic-math-to-today-what-a-year-tells-us", "markdown": "https://wpnews.pro/news/from-gpt-5-can-t-do-basic-math-to-today-what-a-year-tells-us.md", "text": "https://wpnews.pro/news/from-gpt-5-can-t-do-basic-math-to-today-what-a-year-tells-us.txt", "jsonld": "https://wpnews.pro/news/from-gpt-5-can-t-do-basic-math-to-today-what-a-year-tells-us.jsonld"}}