Google is using AI to answer some search queries.

• Google is testing AI-generated answers in regular search results.
• The experiments are labeled “AI Review” and started about a month ago in the US and UK.
• Remember when Google’s goal was to compile lists of sites that had the information you were looking for? That was a long time ago.

Sure, you’ve heard of AI and tech like ChatGPT. Even a growing number of you are trying to use this technology in your daily life — for fun or school or maybe work.

For the most part, though, you’ll have to explore AI if you want to use it in your everyday life. If you want AI-generated results, you need to use the ChatGPT website or app, or click a specific tab on something like Microsoft’s Bing.

Now that’s changing, at least for some Google searchers like me: On Monday morning, using regular old Google, I typed in “Truman doctrine of the Vietnam War” and returned an “AI review.” Found – What Google didn’t find on another site, but something he wrote himself.

I haven’t spent much time reading post-WWII US foreign policy lately — that’s why I asked Google — but I think that was a pretty good answer, actually?

More interesting than the answer was how Google came up with it.

I knew that Google was scrambling to catch up in the AI ​​product wars, and that it was concerned about the fate of its core search product in a world where links to websites with information would no longer be required – that An AI engine will easily generate The answers you need. And I knew that Google was working on an AI-powered version of Search, which you could use yourself.

But I didn’t know that Google had started putting this stuff out routinely, mixed in with every other search result.

Turns out they have about a month. And it may be that Google hasn’t made or said anything big about it — so far, the only place I can find on the web that knows about it is this from the trade journal Search Engine Land. There is an update.

Conclusion: For now, Google is testing self-generated “AI review” for some regular search queries where it thinks the answer might be complex (but responsive). A Google representative told me that AI answers are deployed in a “very limited percentage of search traffic” in the US and UK.

And I shouldn’t be that surprised, really: Ever since OpenAI started blowing people’s minds with ChatGPT in late 2022, it’s been clear that this technology is coming to find out – it’s Microsoft’s with OpenAI. That was the whole point of the big partnership. Technology up to Bing.

But knowing it is one thing and starting to see it in the wild is another. And starting to see it as a general outcome as opposed to something specific.

Google seems to be handling this well, and is fixing many of the obvious problems that Google’s written answers will raise in search.

For example, it clearly labels results as AI-generated, and experimental. And the “learn more” link brings you to this well-written explainer that says “Generative AI is not human. It cannot think for itself or feel emotions. Pattern Finding It’s great to do.” (And, on the downside, “because creative AI is experimental and a work in progress, it can and will make mistakes.”)

And Google shows its work, too: If you click a button in a result, it will display links to helpful, relevant sites like the National Archives.

Because I’m a responsible journalist, I also asked Google how Google Magi, which powers these results, does or doesn’t interact with the technology that powers Google Gemini, its much-maligned app. “wake AI” is a chatbot. I didn’t get a very satisfactory answer, other than that Google would like you to think of them as separate products.

I’m happy to let other people worry about Google’s awareness (you hear a lot less whining about that now, don’t you?). I’m more interested in how these kinds of answers will accelerate the way we already use Google—as a one-stop shop for answers, rather than a place you have to go to anywhere else. Help you find the place where your answers are.

This trend has been well-documented for years and revolves around Google deciding that instead of hanging out at Google—its “knowledge panels” or something similar. By — go elsewhere to get the answer. Although Google’s business model revolves around selling links elsewhere.

Right now, Google tries to keep you on Google. By surfacing text from a site that purports to answer your question. In theory, if you want to know more, you can click. But often, Google’s quote gives you no incentive to click. You have found what you need. You’re done.

(Meanwhile, I can think of many cases where the information Google presents in snippets of text wrong – At least in part because Google is relying on high-ranking web pages whose primary goal is not to be accurate, but to be ranked high by Google. Ask Google, for example, “What is Jason Kelce’s net worth,” and it will feature a (not remotely helpful) answer from a debit card site that also offers search bait about Taylor Swift’s cat. Is.)

And once you start imagining Google providing such fully AI-generated answers all the time, to all kinds of questions, things get really interesting.

On the one hand, Google might become even more valuable because you’re no longer even pretending it’s a “search engine” — it’s just an answer machine. And you go there because you are used to going there.

On the other hand: In this scenario, Google definitely won’t. Only The answering machine means the entire empire, and many businesses that depend on the empire (like, um, digital publishers?) get caught.

You can see why Google is carefully and quietly testing this stuff—and why it needs to find an answer fast.