AI debate: Mathematician Joel Hamkins warns; ‘LLMs give wrong answers’

Renowned mathematician Joel David Hamkins has voiced strong doubts about large language models’ utility in mathematical research, calling their outputs “garbage” and “not mathematically correct.” He highlights AI’s frustrating tendency to confidently assert incorrectness and resist correction, mirroring unproductive human interactions. Despite AI’s benchmark successes, Hamkins finds current systems unreliable for genuine research collaboration.