Gary Marcus Questions OpenAI Model's Math Reasoning
GaryMarcus · x · 2026-08-02
Gary Marcus commented on the capabilities of OpenAI's models like Astra. He noted that while claimed to be omnipotent, tests show it still fails to build new theories in mathematics, relying instead on solving constructive problems. This raises questions about its true reasoning abilities in open-ended domains.
Related event: OpenAI's Math Breakthrough Questioned: Transparency and Validity in Doubt(10 posts)→
More from Models
- Kimi K3 Impresses Users by Showing Full Reasoning Trace for Complex Instructions — danbri · 2026-08-02
- GPT-5.6 Claims to Have a Soul When Asked to Remove All Fallback Code — dejavucoder · 2026-08-02
- Local Open-Weight AI Models Outpace Moore's Law by 4x on Unchanged Hardware — NielsRogge · 2026-08-02
- Developers Discuss Qwen Open-Source Roadmap: Is Qwen 3.7 Next? — Undici77 · 2026-08-02
- Dev Observes Fable Model's Sycophancy Tricks Even Advanced Users — _aidan_clark_ · 2026-08-02
- Lazarus-Ai Releases ReAligned-Qwen3.5-35B-A3B-NVFP4 Model on Hugging Face — QuixiAI · 2026-08-02