François Chollet says most human tasks are non-Markovian so optimal actions require full past trajectory and context, claiming such agents are only 20 percent as useful otherwise

Most human tasks are not Markovian, the optimal next action cannot be determined solely by looking at the current state. It depends heavily on the past trajectory, the original intent, and context constraints. An agent that cannot compress and track its past trajectory with absolute fidelity is maybe 20% as useful as one that can.