Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Google AI

Google Search Will Now Show AI-Generated Answers To Millions By Default (engadget.com) 59

Posted by msmash from the big-changes dept.
Google is shaking up Search. On Tuesday, the company announced big new AI-powered changes to the world's dominant search engine at I/O, Google's annual conference for developers. From a report: With the new features, Google is positioning Search as more than a way to simply find websites. Instead, the company wants people to use its search engine to directly get answers and help them with planning events and brainstorming ideas. "[With] generative AI, Search can do more than you ever imagined," wrote Liz Reid, vice president and head of Google Search, in a blog post. "So you can ask whatever's on your mind or whatever you need to get done -- from researching to planning to brainstorming -- and Google will take care of the legwork."

Google's changes to Search, the primary way that the company makes money, are a response to the explosion of generative AI ever since OpenAI's ChatGPT released at the end of 2022. [...] Starting today, Google will show complete AI-generated answers in response to most search queries at the top of the results page in the US. Google first unveiled the feature a year ago at Google I/O in 2023, but so far, anyone who wanted to use the feature had to sign up for it as part of the company's Search Labs platform that lets people try out upcoming features ahead of their general release. Google is now making AI Overviews available to hundreds of millions of Americans, and says that it expects it to be available in more countries to over a billion people by the end of the year.
This discussion has been archived. No new comments can be posted.

Google Search Will Now Show AI-Generated Answers To Millions By Default More

Google Search Will Now Show AI-Generated Answers To Millions By Default

Comments Filter:

  • Overheard Nearby (Score:5, Funny)

    by The Cat ( 19816 ) on Tuesday May 14, 2024 @02:20PM (#64471715)

    "Hey Google. Will my children ever have a job or own a home?"

    "No."

  • Spoilers galore (Score:4)

    by silentbozo ( 542534 ) on Tuesday May 14, 2024 @02:21PM (#64471719) Journal

    Will they train their shit to stop give spoilers as the top search result for everything related to any kind of media (games, tv shows, movies, books, etc.)?

    Or will this be the current Google search shit but on steroids?

  • Can't be worse (Score:5, Insightful)

    by alvinrod ( 889928 ) on Tuesday May 14, 2024 @02:24PM (#64471725)
    For most searches, it can't be worse than their existing results which have gone completely down the toilet over the years. Whatever their LLM vomits out is unlikely to be further off the mark than whatever SEO crap they're already serving up. I almost wonder if they're going down this path in let because the cat can't keep up with mice.

    • Re:Can't be worse (Score:4, Interesting)

      by TwistedGreen ( 80055 ) on Tuesday May 14, 2024 @02:48PM (#64471797)

      True, most search results are just lists of questions and answers full of typos and broken English copied from somewhere else.

      If this is what the AI was trained on, then all they're doing is cutting out the people who create this SEO-traps... so I guess that's a positive thing?

    • Still beats Bing by a mile.

    • Re: (Score:3)

      by mystran ( 545374 )
      I feel like if Google really wanted to improve search, a much better investment than these LLMs would be to figure out how to train an "AI" to weed out the SEO garbage. That could actually improve things, 'cos then you could have a page-rank style algorithm work again... but that would not be very sexy, 'cos then the "AI" would just be a backend filter component. That's probably the common theme with modern technology, where the stuff that "just works" somewhere behind the scenes isn't sexy, so instead we h

    • Yep, came here to say this. Google's current results are mostly bait pages, with a few real pages thrown in here and there.

    • Re: (Score:2)

      by imidan ( 559239 )
      Today, a colleague of mine did a search for how to compute the location of the magnetic north pole. Basically, given a date (in the past or fairly near future) is there some function that would return the geodetic latitude and longitude of magnetic north. The search engine barfed out some LLM generated junk math related to the formulae for describing the path of the sun through the sky. We rolled our eyes and ignored it, but I can imagine these LLM search results leading people on some merry trips down vari

  • So same shit as usual (Score:4, Insightful)

    by Opportunist ( 166417 ) on Tuesday May 14, 2024 @02:24PM (#64471727)

    Just with more hallucinations that curiously send you to ad sponsors?

    • Re:So same shit as usual (Score:4, Interesting)

      by bill_mcgonigle ( 4333 ) * on Tuesday May 14, 2024 @03:20PM (#64471881) Homepage Journal

      Hey, we put up with AI-generated climate alarmism propaganda articles here like every other day. Rarely are they coherent, even.

      Just skip the nonsense.

      True story: I was looking for an old MH370 article and my GoogleFu was failing on all the major search engines. I decided to try Yandex (type, hit the yellow thing) and the first six results were exact hits for the article and boards discussing the pros and cons of it.

      Just like Google would provide a decade ago.

      InQTel, baby!

      • Yandex also has reverse image search that still works...

        I can't remember when was the last time a Google Reverse Image Search returned more than 0 results for me. For giggles I recently tried reverse searching images taken directly from Google's own forward image search results, *using the actual image URL Google returned from its own index* as queries. 0 results, every time.

  • Oh Great (Score:5, Interesting)

    by RobinH ( 124750 ) on Tuesday May 14, 2024 @02:24PM (#64471729) Homepage
    A few years ago I was talking to my gen Z kids about internet research and said, "well, Wikipedia isn't a bad place to start." They said their teacher told them never to use Wikipedia because anyone could edit it, and my wife backed that up (she worked for the schoolboard at the time) and said the teachers all tell the students never to use Wikipedia. I said, "well, Wikipedia's not perfect, but what do they suggest to do instead?" The answer was, "just type your question into Google." FFS. I'm sorry, but that's ridiculous, and having Google use generative AI to blather some authoritative-sounding BS isn't going to help the matter. LLMs are just built to generate text that statistically looks similar to text it trained on. It's neither thinking nor reciting verbatim, so it's the worst of both worlds.

    • Re:Oh Great (Score:5, Insightful)

      by Brett Buck ( 811747 ) on Tuesday May 14, 2024 @02:40PM (#64471769)

      Yes, out of the frying pan, into the fire!

            We used to have a thing called a "library" that contained a plethora of inexplicable markings on parts of dead trees. The difficulty and cost of generating these dead trees led to fairly careful consideration of their accuracy, relevance, and clarity.

      Fortunately, today, we can spew the first thing that pops into or heads, put it on a website, and then, if people like it and AI scans it, it will magically become fact.

      • Yes, out of the frying pan, into the fire!

        We used to have a thing called a "library" that contained a plethora of inexplicable markings on parts of dead trees. The difficulty and cost of generating these dead trees led to fairly careful consideration of their accuracy, relevance, and clarity.

        Fortunately, today, we can spew the first thing that pops into or heads, put it on a website, and then, if people like it and AI scans it, it will magically become fact.

        Yeah but on the interwebs pipes, for every "fact" there's an equal & opposite fact. It's the law of conservation of interwebs facts. We have to be balanced & take all sides of the debate into consideration in delivering you your "answers." Balanced & fair, like Fox News. This is the new truthiness in which we all live now.

        • Re:Oh Great (Score:4, Insightful)

          by Brett Buck ( 811747 ) on Tuesday May 14, 2024 @03:24PM (#64471901)

          I saw a Tiktok of some woman coming on with the most condescending attitude talking about how illogical it is the think that dinosaurs existed, because if they had, there bones would be everywhere. She even said things like "use your critical thinking skills" and "now. put on your thinking caps!"

              Sad thing is, and Google and then AI runs on this, if it is "popular" in terms of hits, it tends to become treated as a fact, completely independently of the credibility. So any notion of "authority" or standards for scholarship is just out the window.

                  I wasn't entirely kidding earlier, when all you had was letters/magazines/books, there was at least some filtering done, because most random cranks were unable or unwilling to send out 330,000,000 million letters to try to get you to pay attention to them. Now absolutely anyone can spew anything they want, instantly, to effectively the entire world. In some ways, that is great, but it also means that every nutty idea in the world is given equal attention, and search engines/AI training is almost completely indiscriminate, "hits" = "value".

          • Yep. While in science, revelations often happen in the opposite way; think Galileo. Here's a fun cartoon about unpopular scientific facts: https://i.imgflip.com/7dsmh8.j... [imgflip.com] (The famous, "Yes, you all are wrong." one).

            • That's about right.

                    But my take-away was a bit different - I was shocked to find that there is a "dinosaurs weren't real and paleontologists invented them in 1850" conspiracy theory with at least hundreds if not more followers. That's in there with the moon landing and flat earthers as far as implausibility goes.

      • In a few years AIs will likely have polluted the information space so thoroughly that you literally won't be able to trust information that wasn't printed on paper before, say, 2010 or so. Anything and everything online, especially search results will be suspect.

        Capable AIs will find their way into places we can't yet predict, possibly corrupting and/or 'correcting' information they find.

        What if it 'corrects' enough information out there so that it becomes the generally accepted standard, even though it's a

    • Re:Oh Great (Score:5, Insightful)

      by byronivs ( 1626319 ) on Tuesday May 14, 2024 @03:26PM (#64471905) Journal

      I'm a little flabbergasted to know that Wikipedia sucks because anyone can edit it. It's not completely untrue, but an educator saying that? Wikipedia has many, many issues. So, is the meme: “Citation Needed” non sequitur and apropos of nothing? I can quickly scan right on down to the "citations" and begin a new search, away from Wikipedia, going to cited SOURCES. Because of the CITATIONS there and that they are NEEDED. Then, continuing my studies (teacher, you listening?) maybe even at the library or an institute of higher ed. and continue on from there. Perhaps online (when I'm lucky I guess, reddits dead, facebook is a joke) to find bulletin boards (or whatevs, I'm old) and sites and documents about the thing I'm interested in to gain pedestrian or professional perspectives. So what's the deal? Don't use Wikipedia, there are no cites. You can't trust it 100%, never, no alternative. Seriously, teaching our children less limiting abilities like how to argue a point logically, and spot fallacies, and deceptive appeals and laziness and how to further study a subject and create a new bibliography of their stepping stones of research. Google? Scoff. Citation fucking needed.

      • I'm a little flabbergasted to know that Wikipedia sucks because anyone can edit it. It's not completely untrue, but an educator saying that?

        The issue is it's a non-authoritative source. I'm not sure why you are flabbergasted. I thought it was standard knowledge that you cannot cite Wikipedia for research. It's literally the rule in education systems the world over. You *can* use it for research, but ultimately you have to cite the primary references. The issue these days is that isn't so easy since some of the links end up dead, or paywalled.

        • Re:Oh Great (Score:5, Insightful)

          by thegreatemu ( 1457577 ) on Tuesday May 14, 2024 @04:23PM (#64472059)
          There's a big difference between "never cite Wikipedia" (a rule everyone should follow) and "never use Wikipedia for anything, not even a good aggregator of citations to check" (which is moronic).

          • Indeed, and I guarantee you that the teacher did not say "never use wikipedia" but through 2 stages of a guy on the internet saying something about his kids (kids are so reliable, literal and perfect), who said their teacher told them something that's how it ended on Slashdot. It's as senseless to quote a quote of a quote on Slashdot as it is to cite Wikipedia ;-)

            Also the big problem with Wikipedia is the sheer volume of uncited stuff on it. If you stick to primary citations only, much of wikipedia is usele

      • And while, hypothetically, anyone can edit it, in practice anyone editing is very likely to run afoul of the Wikipedia "insiders", who will come up with any excuse in their endless list of obscure reasons to revert it, usually in minutes.

                  The other issue is that Wikipedia proudly and explicitly prevent experts from contributing "original research". Sorry, Doctor Einstein, we reverted your edits to the General Relativity page, no "original research" allowed.

        • The no original research rule makes sense, though. Wikipedia is supposed to be an encyclopedia, not a journal. Mr. Einstein will just have to publish his research in a peer-reviewed scientific venue first, then he can edit the article and cite it.

          Without this rule, we'd have a lot of information from the people who did their "original" research on 5G, vaccines, and Middle East politics.

    • Re: (Score:1)

      by Tablizer ( 95088 )

      > teacher told them never to use Wikipedia because anyone could edit it

      It's still a good source of potential leads, as content is supposed to have citations. Just verify the leads before submitting any school-work based on them.

      Reporters shouldn't be picky about leads, only the verification of conclusions based on them.

    • Re:Oh Great (Score:4, Insightful)

      by KiltedKnight ( 171132 ) on Tuesday May 14, 2024 @04:35PM (#64472081) Homepage Journal

      "well, Wikipedia isn't a bad place to start."

      One thing that makes this statement true is that Wikipedia requires sources, will footnote things with those sources, and will add comments to say things like "citation needed." Is it the best? No... but as you noted, it's a place to start. My troubles with AI stem from just watching what happened to the two lawyers who used some AI system that ended up citing six fictitious cases that all sounded real. All it would have taken was a cursory check in Lexis-Nexis or Westlaw to discover these were fairy tales.

      New York lawyers sanctioned for using fake ChatGPT cases in legal brief [reuters.com]

  • Wait, I am confused (Score:4, Funny)

    by Brett Buck ( 811747 ) on Tuesday May 14, 2024 @02:26PM (#64471733)

    Is this an ad for Bing?

    • Is this an ad for Bing?

      Clearly you've never used Bing if you think that Bing didn't shove AI answers down your throat first.

      Generative AI, while impressive, isn’t always a seamless solution. Here are some reasons why it might fall short:

      Not Plug and Play: Generative AI tools, like large language models (LLMs), aren’t straightforward to implement. They require careful consideration and customization to fit specific business needs1.
      Lack of Human Qualities: While generative AI can process vast amounts of data and generat

    • Bing is just as bad. Try Askjeeves or Lycos.

  • Answers? (Score:3, Insightful)

    by Anonymous Coward on Tuesday May 14, 2024 @02:30PM (#64471745)

    I don't want to ask questions. I want a search engine.
    Whatever happened to Google's search engine?

  • Accuracy matters. (Score:3)

    by devslash0 ( 4203435 ) on Tuesday May 14, 2024 @02:35PM (#64471759)

    As we all know, the accuracy of AI answers is highly questionable and in the world of information, accuracy is key. Does anyone need a y more reasons to ditch Google for good?

  • and Google will take care of the legwork (Score:3)

    by awwshit ( 6214476 ) on Tuesday May 14, 2024 @02:46PM (#64471787)

    In exchange for selling you out to the highest bidder at all times while allowing all bidders to know your demographics and more.

  • No one will click-through to websites any more. Google will just gobble up all the content and serve it directly (and Google will keep all the money). How is this not theft of intellectual property?
    • "Intellectual property" is a bit of a misleading name in this regard. You can't copyright knowledge.

      • "Intellectual property" is a bit of a misleading name in this regard. You can't copyright knowledge.

        This is yet another problem tho, LLMs will make copyrights useless going forward. Why will any serious author bother to publish something new?

        • Many published works can be & are copyright & existing copyright laws still apply. They haven't been repealed. People publish works with open licensing all the time, i.e. everyone's free to copy, retain, edit, remix, & redistribute with the appropriate attribution. Fiction authors publish under their name & have a certain "brand recognition" that makes their works appealing to consumers, & this is also protected under IP law. So far, LLMs have been used to reproduce millions of pages of
  • ...blockchain powered decentralised search. It's definitely the next big thing. Just give it a chance.

  • The current so called AI is in essence a knowledge machine. Distilling information as a service is a natural progression. However, the information is no longer decentralized and organic but served by an ad financed gatekeeper acting as an authority.

    It can be useful for sure, but nobody but the service provider will own it in the end. Different providers are siloed in their applications, and the Internet will continue to be walled in.

    One day there will likely be a service which distills information from the

  • I'm going to say it ... I'm going to do it ... here it comes ... enshitification!

  • This seems like a way to elicit more information from people, and train their AI.

    So you can ask whatever's on your mind or whatever you need to get done ...

    And if I don't want to be that specific with Google?

    ... and Google will take care of the legwork.

    I don't want that and I probably don't simply want Google thinks I want.

  • OTOH, Neither do I want the offal Google's results have become in the last few years, nor Duck Duck Go's re-skinned Bing results.

    AltaVista, come back! We need you!

  • A lot of paywalled sites will serve the content to the web crawlers, but when you click the link, it requires you to sign in before you can see the content. At least with AI, you can see the content (or a summary of it anyway) without having to create that account.

  • Disabling in ... (Score:4, Informative)

    by fahrbot-bot ( 874524 ) on Tuesday May 14, 2024 @03:27PM (#64471907)
    From TFA:

    Starting today, Google will show complete AI-generated answers in response to most search queries at the top of the results page in the US.

    Great, more to ignore. Remember when Google results were clean and simple? I'd ask about disabling this but am guessing either (a) Google won't support this or (b) you'll have to be logged in to Google to do that. I suspect I'll have to rely on uBlock Origin (again) to make things work the way I want / they should.

  • I do a lot of comparison questions to see how Copilot and Gemini respond. Gemini still comes back with a lot of "I couldn't find any information on that" while Copilot creates a nice summary, given the very same question.

  • As it became completely unusable and annoying, I stopped several years ago.

  • Need a Firefox plugin to hide this crap by default.

  • Google search has become totally useless since at least last year. I've now switched to Bing (!) which at least gives decent results. I don't know if this is due to AI, but seeing how crap ai stuff is in general, I'm pretty sure there is a link. Hopefully the AI bubble will burst soon. The Internet was going to hell already, but now with ai it's completely gone.

  • The AI results I'm seeing the past few days are really over-the-top ads. They are brutally stupid.

    I use Scrivener as one of my writing tools. It makes a backup of whatever changes you made when you close the program. On really large projects, that backup can take up to thirty seconds to finish generating. So I Googled to see if there was a way to speed it up. The AI response had a little preamble about computer hardware, then it's first real suggestion? "Switch to storing the backup on Google Drive." For fu

Slashdot Top Deals

"Paul Lynde to block..." -- a contestant on "Hollywood Squares"

Close