> we now know that playing chess well does not require AGI; it is a specific task that can be solved with task-specific algorithms. Hmm. Only if "discrete-tim

> we now know that playing chess well does not require AGI; it is a specific task that can be solved with task-specific algorithms. Hmm. Only if "discrete-time, alternating-games" is the specific-task. Everything within that can use the same algorithms, just with different training data.