--- title: "Google Admits Its AI Search Feature Is a Dumpster Fire, Says It Will Scale Back the Tool" description: "  Google is at least partially admitting that its AI Overviews search feature is a hot mess. The company's generative AI-enhanced search tool has been giving users incredibly dumb answers, confidently telling one user to add about 1/8 cup of non-toxic glue\" to their pizza sauce to stop it from sliding off the pie. Now, […]" date: "2024-05-31" modified: "2024-05-31" authors: - name: "Victor Tangermann" job_title: "Senior Editor" link: "https://futurism.com/authors/victor" url: "https://futurism.com/the-byte/google-admits-ai-search-feature-dumpster-fire" categories: - "Artificial Intelligence" - "Google" tags: - "generative ai" - "Google" - "the digest" --- # Google Admits Its AI Search Feature Is a Dumpster Fire, Says It Will Scale Back the Tool ![Google has released a statement, admitting that "some odd, inaccurate or unhelpful AI Overviews certainly did show up."]() *\Image: Alex Wong/Getty Images\* ## Search Engine Hallucination Google is at least partially admitting that its AI Overviews search feature is a hot mess. The company's generative AI-enhanced search tool has been giving users incredibly dumb answers, [confidently telling one user](), for instance, to add "about 1/8 cup of non-toxic glue" to their pizza sauce to stop it from sliding off the pie. Other users were told to [eat at least one rock a day]() or literally [jump off the Golden Gate Bridge](). Now, the company has [released a statement](), admitting that "some odd, inaccurate or unhelpful AI Overviews certainly did show up." As a result, Google announced that it will limit some responses, especially when it detects "nonsensical queries that shouldn't show an AI Overview." Google maintains that generative AI makes sense as part of its flagship feature — but given the dumpster fire its latest tool is already turning out to be, it still has a lot to prove. "At the scale of the web, with billions of queries coming in every day, there are bound to be some oddities and errors," Google search head Liz Reid wrote in the blog post. "We’ll keep improving when and how we show AI Overviews and strengthening our protections, including for edge cases, and we’re very grateful for the ongoing feedback." ## Sticking Power Some of the observed issues have since been traced back to insincere "shitposts" of Reddit users, suggesting AI Overviews is drawing from some seriously dubious data for its output. The company struck a [$60 million deal with Reddit]() to train its AI on users' posts — and hilariously, the AI is now using Reddit posts to tell people to eat glue. It's not just Google struggling to stuff generative AI into its products with abandon, either. We've already seen chatbots and related tools telling users to [cheat on their wives](), [hallucinate plenty of information](), miserably [fail to summarize existing data](), and make a [complete mess of entire publications](). To be clear, this has been going on for well over a year, and with Google's latest high-profile blunder, it doesn't bode well for the future of the tech. This time, however, is especially egregious considering the prominent position Google Search has held for several decades now. The tool still accounts for the vast majority of the global search engine market, but [early signs point towards]() its competitors making major strides. "We know that people trust Google Search to provide accurate information," Reid wrote. "We hold ourselves to a high standard, as do our users, so we expect and appreciate the feedback, and take it seriously." **More on Google:** *[The Reason That Google's AI Suggests Using Glue on Pizza Shows a Deep Flaw With Tech Companies' AI Obsession]()* ## Author I've been at Futurism since 2017, where my role has evolved to encompass design, writing, and increasingly editing. I've always been fascinated by space exploration and advanced transportation, which I've leaned into by interviewing luminaries in those fields while closely following the dimensions of policy and regulation that allow next-generation projects to succeed -- or, sometimes, to fail. I'm also keenly interested in the effects of generative AI on society, policies, and democratic institutions, as well as clean energy, physics and biology, and the vagaries of tech leadership. My work for Futurism has been cited by publications including Ars Technica, Gizmodo, PC Magazine, Jalopnik, Fox News, and the New York Post. I spent my childhood living in locations including Manila, the Philippines, and Geneva, Switzerland, attended McGill University, and now live in Toronto, Canada. Before Futurism I worked at AskMen and a small photography studio. In my free time, I'm an avid gardener, foodie, and craft beer lover, as well as a maker of artisanal hot pepper sauces. I have a magnificent dog named Freida. ### Author social links [Bluesky]()