4 ms·
We are evaluating the possibility. But this is really for research only. To go beyond that we need further improvements to the safety layer.
by pesenti 6y ago
We are evaluating the possibility. But this is really for research only. To go beyond that we need further improvements to the safety layer.
- aerovistae 6y agoWhat does safety mean in this context? (I'm not an AI researcher.)
- hutzlibu 6y agoMaybe that the bot does not start denying the holocaust, like the one from microsoft. https://en.m.wikipedia.org/wiki/Tay_(bot) https://en.m.wikipedia.org/wiki/Tay_(bot) (but I am not sure if this bot does indeed learn from new conversations)