AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Symbolic Models Do Not Have Any Understanding
"Let's make this as the new, basically, bench mark for evaluating whether or not a system showing any sort of deep understanding," he says. "These models are fundamentally flawed, no question, right? Be talked about this in the last session, that these models are extremely biased and easy to get tricked." He adds: "I personally think that whoever can have claims that these models do not any word model, don't have any kind of human like cognition ... is it really incorrect?"