Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
Google AI

Will AI Kill Google? (yahoo.com) 54

"The past 15 years were unique in ways that might be a bad predictor of our future," writes the Washington Post, with a surge in the number of internet users since 2010, and everyone spending more time online.

But today, "lots of smart people believe that artificial intelligence will upend how you find information. Googling is so yesterday." Sam Altman, the top executive overseeing ChatGPT, has said that AI has a good shot at shoving aside Google search. Bill Gates predicted that emerging AI will do tasks like researching your ideal running shoes and automatically placing an order so you'll "never go to a search site again." In defending itself from a judge's decision that it runs an illegal monopoly, Google says the company might be roadkill as AI and other new technologies change how you find information. (On Wednesday, the U.S. government asked the judge to overhaul Google to undo its monopoly.)

But predictions of Google's looming obsolescence have been wrong before, which calls for humility in fortune-telling our collective technology habits. We're devilishly unpredictable.... Maybe it's right to extrapolate from how people are starting to use AI today. Or maybe that's the mistake that Jobs made when he said no one was searching on iPhones. It wasn't wrong in 2010, but it was within a few years. Or what if AI upends how billions of us find information and we still keep on Googling? "The notion that we can predict how these new technologies are going to evolve is silly," said David B. Yoffie, a Harvard Business School professor who has spent decades studying the technology industry.

Amit Mehta, the judge overseeing the Google monopoly case, formed his own view on AI moving us away from searching Google. "AI may someday fundamentally alter search, but not anytime soon," he said.

Will AI Kill Google?

Comments Filter:
  • by Rosco P. Coltrane ( 209368 ) on Saturday November 23, 2024 @11:35AM (#64966867)

    One can only hope.

    • by Tablizer ( 95088 )

      Finally a good use of AI.

      • I have already replaced search with AI for technical queries. The Gemini support inside of Android Studio is far better than trying to use web search for the same queries. Both methods are searching the web and Gemini includes the URLs for where it found the answers. The key difference is that Gemini understands the context in which the query is being made. For example it knows I am working in Kotlin on Android so it doesn't give me answers containing Java like a web search will. It also knows I am using C

      • by gweihir ( 88907 )

        Indeed. It will probably fail at it though.

    • One can only hope.

      My hope is that AI changes all of these companies, inducing them to offer something better than their current products.

      [Yes, I know many of them make money with ads, so we are the product.]

      • The quality of those company's offering really isn't what's wrong with them.

        Their business model, the very way they exist at all, is a threat to personal freedoms, democracy, and an insult to decency and humanity.

        • by shanen ( 462549 )

          So how do you feel about a progressive tax on corporate profits linked to market share? Various ways to detect when a corporate cancer has started stomping on freedom, but the two most obvious are customers with no other choices and wannabe competitors who are blocked from entering the market.

        • If you hate their services, why do you use them? No one is making you use them.

          Why do people think they have a right to use all of these corporate offerings in complete anonymity for free? Corporations aren't charities. These services aren't free to provide, either pay with information, pay with cash or don't use them, pick one. If you want complete anonymity you're going to have to fork over some cash.

          • And please don't go off on, why do I have to pay for privacy??? You AREN'T, you are paying to use the service. You chose not to allow use of your data as an alternative form of payment so you will have to use good old cash.

          • why do you use them? No one is making you use them.

            Because more often than not, there's no other practical choice.

            You almost need a cellphone to lead a normal life now. Well, for that, you need to submit to Google's or AppleÂs surveillance.

            You almost need the internet too. They're all there to watch you too. And they're not even offering any product most of the time: they just watch you constantly in the background. If you don't want to submit and you block the surveillance, more often than not, the services you need to access on the internet break.

            You

            • Wow, I managed to make it through the first 30 years of my life before all of this was invented. So that proves it can be done. And you really can still mail checks to places to pay your bills. Telephone works too for calling people or you can write them letters. You are choosing to participate, it is not forced.

            • Today I was trying to find out if a specific cutoff disk (on Amazon) would fit my angle grinder, so I googled and came up with this link [thehandybee.com].

              Note from the otherwise useless article:

              The angles on an angle grinder disc can also be interchangeable depending on the model of the angle grinder you have. The best way to check if your angles are interchangeable is to get a feel for the size of your tool and then experiment with different types of discs until you find one that fits your machine.

              Obvious hallucination written by an AI, but most people wouldn't be able to tell.

              It's getting much harder to avoid the AI crap any more.

              I'm spending more and more of my search time in google just trying to avoid crap and getting to the actual answer to my query.

    • by shanen ( 462549 )

      Modify parent Funny even though it was the obvious low-hanging fruit joke I was looking for.

      Sure tempted to rant about solution approaches to the corporate cancer problem, but on Slashdot solutions have become the biggest jokes.

    • Won't kill FB as real people are complementary to AI. You can't fully replace one with the other. On the other hand, LLMs hurt search much more. And Google has no access into FB, while FB has access to the web.
  • "No."

    First, there is no AI, just regurgatitive crap.

    Second Google is fueled by imagnative human intelligence.

    Third, Betteridge's law of headlines says no.

    • AI could be a great improvement to search. I don't quite trust it enough to just give me an answer to a qustion, but it would work great for semantic search, i.e. a search engine that understands different meanings of words, and can ask you further questions to refine your search. I've played with an early (non AI) prototype of this and it worked pretty well. Using AI to not just interpret results but also interpret your query, it could skip through the initial back & forth to establish context, and
      • I have found AI to be somewhat conciliatory: too eager to placate the questioner, and unwilling to push back on stupid questions.

        I hope AI will not create a population of self-affirming conspiracy-theorists. Sure, perhaps that has already happened, but AI could make it much worse if there's an implied imprimatur for something AI spits out.

        Then again, maybe AI (or more generally, data analysis) could help. See Ground News [ground.news] for example.

        [I have no ties to Ground News. I just think their technology is cool.]

      • by cowdung ( 702933 )

        There was no moment in history that Google search didn't use AI.

        Google is and has always been an AI company.

        Now, maybe LLMs (which I think is what you mean by AI) could improve search. Though the jury is still out on that one.
        Google has been using Neural Nets since forever. And invented the Transformer.

        I thought people in Slashdot where technical. I guess I was wrong.

        • In common parlance, AI = having the semblance of intelligence and.or creativity. LLMs, GANs and VAEs and the like. Google search? Not really. Also I was under the impression that Google search was mostly algorithmic in nature rather than driven by neural nets or other AI-like tech.

          I was under the impression that Slashdot readers were mostly pedantic... seems I was right.
      • Google has AI responses to natural language questions. They appear "sometimes". Sometimes they are very useful. Sometimes they are complete horseshit. Either way they come with links which you can read. It's pretty good at giving technical answers to questions, not as good at giving answers to technical questions. If you don't know anything it might really confuse you.

        The results are marked as 'experimental'.

    • No clear bullet-proof definition of AI exists. Been in zillions of debates on that. Stop already. Colloquial World calls it "AI", deal.

      • by cowdung ( 702933 )

        Of course there are very clear definitions of AI. Here is what Norvig and Russell say in their classic book "AI: A Modern Approach":

        “the study of [intelligent] agents that receive precepts from the environment and take action. Each such agent is implemented by a function that maps percepts to actions, and we cover different ways to represent these functions, such as production systems, reactive agents, logical planners, neural networks, and decision-theoretic systems”

        Now, in popular culture, peo

    • > First, there is no AI, just regurgatitive crap.

      You heard it folks, pack up and let's go home. AI is not original. Why does it need to be original, we have humans prompting it. It only needs to assist well.

  • I vote for the equivalent of a matter-antimatter collision, destroying both and leading to a massive release of free (no longer used) datacenters.

  • Enshitification (Score:5, Informative)

    by 0xG ( 712423 ) on Saturday November 23, 2024 @12:06PM (#64966903)

    The thing most likely thing to kill Google is Google itself.
    - Stop with the AI-generated summaries, which are often very incorrect
    - Stop prompting me to log in
    - Stop prompting me to enable javascript
    - Stop prompting me to enable cookies
    - Stop prompting me to use chrome
    - Stop telling me 'it's better with the app'

    Just give me plain old search results.

    • Stop with the AI-generated summaries, which are often very incorrect

      I've been astounded at how terrible the AI summaries on Google have been. I realize that they probably can't afford to have their highest-performing LLMs generating answers for every Google search, but these summaries are so terrible that they risk poisoning the public's opinion of their overall AI capabilities.

    • by gweihir ( 88907 )

      I recommend DuckDuckGo. Google exceeded my annoyance-threshold a few years back and since then I use it perhaps once a month to confirm that it does not find something either.

  • by timeOday ( 582209 ) on Saturday November 23, 2024 @12:15PM (#64966921)
    Right now I'm having a much better experience working with an ongoing side discussion with ChatGPT for ideas, tradeoffs, and mechanics (like function arguments etc).

    google could match that, but won't, because ads are its lifeblood. The 'search for web pages' experience has become so encumbered with highly-profitable advertisements, cookies and other hassles that it's hard to separate the inherent advantages of conversational AI from the advantage it enjoys simply by not being junked up yet.

    But can AI survive in that fashion? I doubt it. It's free-riding on all the ad-supported website content that it scraped.

    Let's hope conversational AI competitors will force google to meet it partway, and dial back the crap somewhat. And let's hope that OpenAI etc doesn't completely go to shit, like the web.

    • > But can AI survive in that fashion? I doubt it. It's free riding on all the ad supported website content that it scraped.

      You are missing a big piece. LLM-human chat rooms. They serve 330M users (300M openAI and 30M Anthropic). These people feed data, documents, specifics into the model, and a lot of feedback. The model can just stay with its mouth open and data falls in. Interactive, task data, not web junk. Humans provide testing, implementation and feedback. This sessions refine the model so it le
  • Google killed Google (Score:3, Interesting)

    by Anonymous Coward on Saturday November 23, 2024 @12:29PM (#64966935)

    More specifically, they hired the former head of Yahoo Search [wheresyoured.at] to lead the Google Search team and turned the Google search results into the shitshow of ads and sponsored placements that are seen today. He was quietly retired about a month ago [wheresyoured.at], but the damage is done. AI is just accelerating Google's demise.

  • by michael_cain ( 66650 ) on Saturday November 23, 2024 @12:32PM (#64966941) Journal

    On Wednesday, the U.S. government asked the judge to overhaul Google to undo its monopoly.

    Those of us who are old remember the late 90s when the US government got farther than that, and the judge in the Microsoft antitrust case had already decided to break up the company. Then Bush appointees replaced Clinton appointees and the government abruptly asked the judge not to do that. I really don't think Google has to worry about being restructured.

  • by Casandro ( 751346 ) on Saturday November 23, 2024 @12:53PM (#64966959)

    Remember when Google was actually good, when it had relevant results up front. Before that you had search engines where the result you were looking for was in down 10 pages of results or more.

    Today you search for something, and Google just ignores what you entered, and displays irrelevant pages, often with no relation to the words you entered. It is no longer a full-text search, but something much worse.

    LLMs are not good at storing information, they are both inefficient and lossy.

  • by organgtool ( 966989 ) on Saturday November 23, 2024 @12:54PM (#64966961)
    I was a bit late to the LLM party, but Google's poor performance has changed that in a big way. SEO has gotten so bad that I rarely find what I need on the first page of Google results. And when I do find a page that contains the answer, I often need to read through most of the thread before I find the answer to my question. At this point, it's just way faster to query an LLM. Even if it hallucinates some details, the overall gist is usually close enough that I can do the rest on my own or use a search engine for the last important bits. But due to Google's declining results, I find myself using search engines other than Google to get those finishing details.
    • by jythie ( 914043 )
      So you end up replacing people figuring out how to optimize for google's current algorithm for people figuring out how to optimize for various LLMs instead.
      • That is correct. We are in the golden age right now where AI providers haven't bastardized the output with ads and content providers haven't found a way to game the results. Enjoy it while it lasts!
  • Kinda sounds like something google can just fold into its current search, and shift its business model a bit so instead of selling ads you sell AI routing people right to your product.
  • All I can say is that I don't use google as much as I used to. For coding it was google to search stack overflow. AI solves my issue in the context of my code. Now I only use google if I'm looking for something to buy.
  • Google killed the internet by creating an algorithm that caused everyone to game the system with SEO crap. I hope Google goes bankrupt.

  • ...when they ousted Ben Gomes, inventor of the PageRank algorithm, and replaced him with Google's Head of Ads, engineer class traitor Prabhakar Raghavan. You know, the guy in charge of Yahoo! Search from 2005-2012. You can't make this shit up. https://www.wheresyoured.at/th... [wheresyoured.at]
  • AI is only appears to be intelligent which is to say that it's intelligence is purely illusionary. The problem with this is that the resulting information from a generative AI cannot be trusted to be accurate specifically because what AI lacks is rationality which is what underpins discernment when it comes to processing information. That said, AI is not useless as it may be usable in organizing search results in a better fashion than our current algorithms.

    Ai aside, AGI is differentiated in that it is capa

  • by TheMiddleRoad ( 1153113 ) on Saturday November 23, 2024 @04:52PM (#64967307)

    All those BS sites with great SEO get to the top of the search list, ruining the results. Then all those AI sites train the LLM and give bad results too. It's a cycle into the toilet.

  • by PPH ( 736903 )

    But not fast enough.

  • Google search results (even with google-foo) have been getting worse since at least 2018.

    They got greedy. They sold their truth to the highest bidder.

    Despite the possible inaccuracies, the LLM's with internet access are giving much better search results and answers to factual queries.

    I've found ..

    Decreasing Effectiveness of Advanced Query Techniques due to google trying to shift to natural language processing (and failing), Tricks like using quotes for exact matches, minus signs for exclusions, or specif

"Someone's been mean to you! Tell me who it is, so I can punch him tastefully." -- Ralph Bakshi's Mighty Mouse

Working...