Google’s AI Mode finally reached the UK this week, and I hate it. Open Google in a browser and the ‘AI Mode’ button shimmers, begging to be clicked. And every results page has AI Mode lurking, menacingly, as the first tab. Before Videos. Before Images. Even before ‘All’, which is exactly how Google wants it.
One day, Google will flick a switch. Every query will default to AI Mode. Other tabs will be secondary, then vanish entirely. “But hang on,” you might say. “So what? Isn’t it easier to get instant answers than scroll through pages of links? Isn’t that… progress?” Not entirely, no.
I’m not saying AI search is useless. But its perceived value is way out of step with what’s being delivered. Too often, we’re getting the illusion of knowledge rather than understanding. Chatbots train people to skip the hard bit – the digging, the thinking – and trust in what’s served up, without really knowing whether it’s any good.
Frequently, it isn’t. Remember, Google AI not that long ago recommended you add glue to pizza to stop the cheese from sliding off. It couldn’t tell the difference between a Reddit joke and culinary advice. You might laugh but people have been injured through following the advice of chatbot ‘hallucinations’. The combination of confidently provided results and the baked-in trust people have in search engines is a dangerous mix.
I increasingly see people regularly copy-pasting chatbot answers into forums as if they’re gospel. But systems vying for the role of internet overlord – from ChatGPT to Google AI Mode – don’t understand context. They remix existing content in ways that look plausible as a response to a prompt. And chatbots also feed confirmation bias – buttering users up to keep them happy – more than any list of links.
Source unknown
It feels like we’re blazing into a future where generations will have abandoned – or not learned – critical thinking in research. They’ll accept or even prefer a slick-sounding synopsis of what they might have been looking for over any level of manual graft because it’s a shortcut.
Who cares if it’s wrong? It’s probably close enough. And if it isn’t, well, searching for stuff takes ages. Besides, it’s the future. How long before you can’t search online properly anyway? Sources are already sinking beneath the surface. We’re a step away from them vanishing entirely, leaving everyone with no idea how any AI search result sausage was made. And that… is also exactly how Google wants it. The search engine that was once a window to the web is now threatening to become a screen displaying the output of a black box. And that won’t want you thinking about other black boxes or the open web. Google AI Mode is ‘All’. Probably sooner than we think.
And this is just step one. Next, tech bros want agentic AI tendrils slithering into your entire online world. Today, it’s all about handing over search queries to chatbots. Soon, you’ll be encouraged to offload everything from calendars to purchases.
I think that sounds positively dystopian. Although maybe I’m just deep in the third stage of Douglas Adams’s rules of technology. He said anything that exists when you’re born is normal. Anything invented before you turn 35 is new and exciting. And anything that arrives after that is “against the natural order of things”. Reader, I’m way beyond 35.
But considering Google AI recently also suggested humans eat rocks, I’m not entrusting it with my search results, let alone anything more. So if you need me, I’ll be rocking in a corner, prodding DuckDuckGo and Kagi and praying they too don’t get their own shimmering AI buttons.