2 ms·
I'm reminded of a much more obvious example. https://www.theverge.com/2016/3/24/11297050/tay-microsoft-chatbot-racist https://www.theverge.com/2016/3/24/112970
by jayd16 9y ago
I'm reminded of a much more obvious example.
https://www.theverge.com/2016/3/24/11297050/tay-microsoft-chatbot-racist https://www.theverge.com/2016/3/24/11297050/tay-microsoft-ch...
I see a lot of comments about how its somehow sinister to want your model to be better than the lowest common denominator and that is pretty damn ridiculous.