Imagine this: you type a question into Google, and instead of just links, an AI pulls up a quote from a random Reddit user—and presents it as "expert advice." That's the new reality of Google's AI search, and it's equal parts thrilling and terrifying.
Google is rolling out updates that mine discussion boards and social media posts for first-hand information. Now, your AI responses will feature specific quotes from forums like Reddit, complete with the poster's name or handle, giving you a "preview of perspectives" straight from the digital crowd.
But here's the catch: just because a response is labeled "expert advice" doesn't mean it's trustworthy. AI still struggles to detect sarcasm and humor. Google's own systems have already recycled jokes from Reddit as if they were serious, life-saving tips.
And that's not the only problem. Generative AI is notorious for hallucinating—making up fake news summaries, bogus legal advice, and delivering it all with unnerving confidence. Even if AI Overviews are accurate 90% of the time, that still leaves one in ten responses riddled with errors.
Google is trying to help. Source links now appear directly inside AI responses, and hovering over them shows a preview of the website. The system will also prioritize content from your news subscriptions, flagging information from sources you actually trust.
But don't let that lull you into complacency. You still need to do the extra work: click through to the source material and verify it actually says what the AI claims. Ask yourself—does "user-generated" mean "expert"? Spoiler: usually not.
Use lateral reading: hunt for reputable sources that support or refute what the AI tells you. In a world where AI cheerfully bluffs, your own critical thinking is the only safety net that matters.
