AI Hasn't Solved Math Conjectures Yet, but the Bar is Rising Fast
adonis_singh · x · 2026-08-01
Commenting on the fact that current AI models like Astra haven't yet created new branches of mathematics or posed interesting conjectures, the author points out that the bar for AGI validation has become incredibly high. Despite these current limitations, he has no doubt that AI will eventually cross this threshold.
Reflecting on the recent release of the o3 model, he notes that an enormous amount of progress has happened over the past year, bringing high-level logical reasoning closer than ever.
More from AGI Musings
- Mathematicians Fight Back: Leiden Declaration Targets AI Overreach — johnseach · 2026-08-01
- Scholar Argues AI Agents Should Act as Personal Advocates Against Bureaucracy — sethlazar · 2026-08-01
- The Guardian: China's Open-Source AI Models Upend Silicon Valley and Spark Policy Chaos — nordicinst · 2026-08-01
- AI Math Proofs Mostly Find Examples, Human Mathematicians Still Irreplaceable — JFPuget · 2026-08-01
- AGI is Overrated: We Only Need 'Narrow Scientific Superintelligence' — scaling01 · 2026-08-01
- Turing Award Winner Patterson: Direct Horses-to-Autonomous Cars Transition Would've Been Unquestioned — davidpattersonx · 2026-08-01