A 30-year-old open problem in logic resolved with ChatGPT, disproving the stable forking conjecture
Dr_Singularity · x · 2026-09-03
Researchers report resolving the stable forking conjecture — a major open problem in model theory dating back 30 years — by finding a counterexample with the help of ChatGPT. The result highlights LLMs' emerging role in mathematical research.
Related event: ChatGPT Helps Disprove 30-Year-Old Stable Forking Conjecture(3 posts)→
More from Research
- Training Trick: Fine-Tune Only the Last Layers to Avoid the Moving-Target Problem — Sauers_ · 2026-10-18
- Models Can Be Trained to Sense Activation Probes and Generalize Across GPUs Zero-Shot — Sauers_ · 2026-10-18
- Models Can Learn to Detect Training vs. Deployment and Generalize to New Runtimes — Sauers_ · 2026-10-18
- How One Attention Head Acts as a Switch to Amplify Tiny Probe Differences — Sauers_ · 2026-10-18
- Model trained to express different preferences only when its activations are probed — Sauers_ · 2026-10-18
- On Qwen3-8B, even a kernel-split without a probe is detectable to a trained model — Sauers_ · 2026-10-18