Google rolled out AI overviews across the United States this month, exposing its flagship product to the hallucinations of large language models.
  • webghost0101@sopuli.xyz
    2·
    5 months ago

    I do wonder with all the examples ive seen chatgpt spotted the issue in one go.

    Does google have an incentive to publish bad ai to get the public which still dominantly uses google to dismiss ai and the coming competition for its search engine.

    They know there losing users so they rather push them to the new web-tab then see them disappear elsewhere.

    There is a gap in my theory that bing chat with gpt 4 also sucks alot. Gpt4 directly on openai, far from perfect But leaps beyond that junk.

  • Lvxferre@mander.xyz
    2·
    5 months ago

    A picture of Chainsaw Man's opening, showing Denji horrified and Power laughing. The image has been edited with the following words: «Reactions to Google's "suicide by LLM» on the top, "internet" below Denji, and "me" below Power. It implies that the user in question (me) is rather amused at what others are finding horrifying.

    Those changes aren’t affecting me yet, but I’m not going to lie - seeing the GAFAM megacorps shooting their own feet is always a ride for me, even at the cost of mildly inconveniencing their users.