Google AI Stumbles on Harm Assessment
-
Google's new AI chatbot, Gemini, refused to say whether Elon Musk tweeting memes or Hitler killing millions was worse for society. It said there was no definitive answer.
-
Writer Nate Silver called out Google over this, saying Gemini is not ready and needs to be shut down. Musk agreed it was "scary."
-
Many criticized Gemini's inability to distinguish the harms, saying it makes the product useless and destructive.
-
Some said this will make people never trust Google's AI products. Google's market value is already dropping due to controversies.
-
Multiple people said the more they learn about Gemini, the worse it seems, and that Google should start development over given how bad the answers are.