You know how Google’s new feature called AI Overviews is prone to spitting out wildly incorrect answers to search queries? In one instance, AI Overviews told a user to use glue on pizza to make sure the cheese won’t slide off (pssst…please don’t do this.)
Well, according to an interview at The Vergewith Google CEO Sundar Pichai published earlier this week, just before criticism of the outputs really took off, these “hallucinations” are an “inherent feature” of AI large language models (LLM), which is what drives AI Overviews, and this feature “is still an unsolved problem.”
I’ve seen suggestions that the AI Overview is based on the top search results for the query, so the terrible answers may be more to do with Google Search just being bad than any issue with their AI. The AI Overview just makes things a bit worse by removing the context, so you can’t see the glue on pizza suggestion was a joke on reddit or it was The Onion suggesting eating rocks.
I noticed that while using phind and perplexity. Its context is vitiated with results from sites that rig SEO, which are almost copy/paste with the same garbage, so instead of answering the question it makes a useless summary of them. Even asking chatgpt usually gives more correct answers.