3 ms·
this whole emphasis that openai and deepmind put on human/robot collaboration is a paper thin pr move in my opinion. the robots will be better than any human, h
by bronz 9y ago
this whole emphasis that openai and deepmind put on human/robot collaboration is a paper thin pr move in my opinion. the robots will be better than any human, humans will not be able to contribute a single thing soon. but they try to make us all feel safe by making it look like they benefit from our brains.
- cglouch 9y agoI know it's not the same type of AI, but in chess there's a whole scene for computer + human play. A chess engine on its own can have trouble seeing strategic ideas that humans can recognize (e.g. opposite-colored bishop endgames, certain closed positions, and fortresses) so an engine on its own will lose to that engine being assisted by a skilled human player. In other words, humans are still capable for contributing at least a little. That said, it's not much - I think a chess GM paired with an engine will probably only be able to beat an engine rated ~100 or so points higher than their own. It will be interesting for deep learning, though, where the ideas are a bit more abstract. Perhaps humans will be useful for a while longer.
- vbezhenar 9y agoAI uses many simulations to train its network. It's appropriate for chess or Dota, but you can't do that for real war, for example, there were not enough wars to learn from them and you can't simulate war good enough. Or making a business plan for Oracle corporation: it's unique situation, you don't have millions of Oracles bankrupting to learn from it. Humanity has this knowledge, but it's encoded in books, teachers and experts. So AI will need to communicate with people to adapt to our society, they can offer advices to experts, but they need to learn from those experts first.