Chess Model Limitations
The discussion reveals the challenges of training models to match arbitrary skill levels in chess, highlighting that simply adjusting Elo ratings does not guarantee improved performance. Nicholas shares his motivations for attacking systems, asserting that his drive comes from the enjoyment of solving puzzles rather than altruistic intentions, which has led to criticism from some in the community.In this clip
From this podcast

Machine Learning Street Talk (MLST)
Nicholas Carlini (Google DeepMind)
Related Questions