Last week, Google launched AI search results for millions of users. The goal was to provide a better search experience. Instead, the AI produced all sorts of weird results, like telling people they should put glue on their pizza to help the cheese stick. eat stone.
Now, a company blog post Published Thursday by Google’s search chief Liz Reid, the tech giant doubled down on “data gaps” due to inaccurate results, asking weird questions with people and claiming AI results lead to “higher satisfaction” from Search. Reid claims that AI Visions generally “do not hallucinate;” they just sometimes misinterpret what’s on the internet.
“There’s nothing like millions of people using this feature with tons of new searches,” he writes. “We also saw pointless new searches that tended to produce the wrong results.” He also correctly points out that “numerous fake screenshots” of AI Views have circulated the web.
First, I would like to mention “which mammal has the most bones?” It’s a fair question if you’ve spent any time with an inquisitive child. Second, Google lured millions of people into the feature, which resulted in a lot of backlash and even prompted articles explaining how to improvise to disable it.
Reid’s blog also explains how Google has tweaked its AI views by limiting what it sees to “nonsense” queries and satire. It’s good to address this because many people were surprised by the popular satirical website Onion and Reddit user “fucksmith” was not filtered out of the AI results in the first place.
Part of Reid’s blog also compares AI Insights to another long-form Search feature called Featured Snippets, which highlights information from a relevant web page without using generative artificial intelligence. According to Reid, the “degree of accuracy” for the snippets provided is “on par” with AI Insights.
If Google is going to compete, it needs to move fast. But it must also protect user trust. After AI Visions told us all to eat Elmer’s glue, that might be hard to recover from.