Bill Gates Says AI Could Kill Google Search and Amazon As We Know Them (cnbc.com) 179
An anonymous reader quotes a report from CNBC: Microsoft co-founder Bill Gates believes the future top company in artificial intelligence will likely have created a personal digital agent that can perform certain tasks for people. The technology will be so profound, it could radically alter user behaviors. "Whoever wins the personal agent, that's the big thing, because you will never go to a search site again, you will never go to a productivity site, you'll never go to Amazon again," he said.
This yet-to-be developed AI assistant will be able to understand a person's needs and habits and will help them "read the stuff you don't have time to read," Gates said Monday during a Goldman Sachs and SV Angel event in San Francisco on the topic of artificial intelligence. Gates said there is a 50-50 chance that this future AI winner will be either a startup or a tech giant. "I'd be disappointed if Microsoft didn't come in there," Gates said. "But I'm impressed with a couple of startups, including Inflection," he added referring to Inflection.AI, co-founded by former DeepMind executive Mustafa Suleyman.
It will take some time until this powerful future digital agent is ready for mainstream use, Gates said. Until then, companies will continue embedding so-called generative AI technologies akin to OpenAI's popular ChatGPT into their own products. [...] He also likened the rise of generative AI technologies that can produce compelling text as a game-changer that will affect white-collar workers. Gates added that he believes that future humanoid robots that are cheaper for companies to use than human employees will greatly impact blue-collar workers, too. "As we invent these robots, we just need to make sure they don't get Alzheimer's," Gates said in jest.
This yet-to-be developed AI assistant will be able to understand a person's needs and habits and will help them "read the stuff you don't have time to read," Gates said Monday during a Goldman Sachs and SV Angel event in San Francisco on the topic of artificial intelligence. Gates said there is a 50-50 chance that this future AI winner will be either a startup or a tech giant. "I'd be disappointed if Microsoft didn't come in there," Gates said. "But I'm impressed with a couple of startups, including Inflection," he added referring to Inflection.AI, co-founded by former DeepMind executive Mustafa Suleyman.
It will take some time until this powerful future digital agent is ready for mainstream use, Gates said. Until then, companies will continue embedding so-called generative AI technologies akin to OpenAI's popular ChatGPT into their own products. [...] He also likened the rise of generative AI technologies that can produce compelling text as a game-changer that will affect white-collar workers. Gates added that he believes that future humanoid robots that are cheaper for companies to use than human employees will greatly impact blue-collar workers, too. "As we invent these robots, we just need to make sure they don't get Alzheimer's," Gates said in jest.
I wish (Score:5, Insightful)
Re:I wish (Score:5, Insightful)
It basically is a search engine, but without any way to tell where it pulled the info from.
I don't think Gates' prediction about "you will never go to a search site again" is that insightful; I mostly already quit using search engines already. Their usefulness has cratered in the past 15 years. Part of that is the signal-to-noise ratio on the internet got way worse. But it's also Google catering to the lowest common denominator user. Boolean search operators were a great way to tune your results to usefulness, and they took it away. More recently (4-5 years ago) the search engine started pulling out what it thinks is a direct answer to my question and highlighting that in a box on the results page. Often it's wrong. It's obviously there to cater to idiots for whom clicking on a search result to see the original source and its context is just too hard.
AI search assistants are a continuation of that downward slide. I get it, most people are idiots, so big companies have to "center them in the experience." But making idiots into more effective, faster idiots is a net negative for the rest of us who have to live with them. I feel like it'd be better to have some kind of very low bar to run an internet search. Like that you have to at least look at different sources, and be given a chance to evaluate them, even if you don't take it.
In practice, AI will be most useful for running scams (scam bots) and will attract idiots like flies to honey. Edge cases where AI makes a positive impact will be few and far between. In 10 years' time, it'll be so bad even normies will talk about how it didn't live up to the hype. Much like Bitcoin in that respect.
Re: (Score:2)
It basically is a search engine, but without any way to tell where it pulled the info from.
Except it also gives you a bunch of boiler-plate nonsense to read as well.
Re: I wish (Score:2)
I loved the "stuff you don't have time to read" part - presumably that's the mindless dross that these "AI" systems are generating.
Re: (Score:2)
AI: Text for tards. Stuff you don't have time to read, stuff you don't have time to type.
Re: (Score:2)
Actually it's not a search engine. A search engine does not make things up. It is exactly as it is described - it is a language model that attempts to generate additional text output based on the token embeddings (words) that have come so far. The fact that the corpus it was trained on contains some useful information means that the output ChatGPT produces can be useful and sometimes factual. But there is zero "understanding" or models of the facts. At its heart it is merely word weights and distribution, a
Re: (Score:2)
Stephen Wolfram has written an incredibly concise and detailed explanation of what ChatGPT does and how it does it [stephenwolfram.com]. I suggest anyone with interest in this sort of thing read this to really understand what is going on, and what the limitations are.
These bits sum it up nicely:
I second that. I have not read the full thing yet, but so far accurate, clear and to the point. The thing is rather long, but even the first few pages give you a good idea as to how limited ChatGPT and family really is.
Re: (Score:2)
People tell me that I’m using google incorrectly because my searches contain keywords. But most times I’m not asking a single question, I’m researching an old piece of hardware and looking for documentation or people mentioning that piece of hardware.
Re: (Score:3)
Who tells you that Google searches with key words are "incorrect"? That is the functional core of all search technology, the use of distinctive terms ("key words") that indicate the relevance of a piece of text. More recent iterations of search engines use information theoretic models of relevance instead of the old vector distance computation, but it is still scoring relevance on those distinctive terms. The idea that Google is really doing something "better" if you type in a question is false, it is a "st
Re: (Score:2)
You know, there is a very simple conclusion here: Gates also is an idiot. That he is one outside of anything business-related he has proven time and again.
A good point. The super-rich are invariably treated with special deference. The recent Glass Onion movie was a refreshing corrective in suggesting that a tech "genius" can very easily be an utter moron out of the narrow range of skills that allowed them to amass wealth. Many "tech geniuses" have more than a touch of Miles Bron - Musk obviously, Gates, Thiel, etc.
Re: (Score:2)
I'd like to have a better search engine, but the ChatGPT needs to solve the problem of "confidently explaining completely false information" before it can replace a search engine.
Why does it need to solve that problem BEFORE it can replace a search engine? I think that 80% of my existing search results on google get confident explanations of wrong information already.
Re: (Score:2)
Why does it need to solve that problem BEFORE it can replace a search engine?
Mainly because it presents itself as authoritative when it is not. Search engines don't present themselves as authoritative (although I admit sometimes in the "questions" section Google presents itself as authoritative when it tells you completely false things).
Re:I wish (Score:4, Informative)
ChatGPT authoritative?
It's a chatbot?
I never understand why people have this expectation of perfection for AI.
The whole point of AI is dealing with imprecision. ChatGPT is at best like your buddy. You can talk to it. It may give you some good conversation. But it may also assume things, make up things, and go off on a tangent.
I don't get why people expect it to act like some academic scholar. Do most your beer buddies give you references including page numbers every time they are talking about some topic?
Re: (Score:2)
ChatGPT is at best like your buddy.
No. My "buddies" have a decent sense of when they know something. And usually they do.
Re:I wish (Score:5, Funny)
If ChatGPT can automatically append "-quora" on to every Google search I make, I may think of switching.
Re:I wish (Score:4, Interesting)
Before the internet, if you wanted to know something you had two choices. You could ask your aunt who always seemed to have the answers, or you could go to the library.
Turns out your aunt just confidently told you whatever nonsense she had been told as a child, and actually knew very little. The library was better, but many books were either out of date or contained inaccuracies, deliberate or accidental.
Now we have an even more dangerous situation where people think they can get the truth from the internet, so they "do their own research" (google search and reading some Facebook posts by their aunt) and end up believing that COVID vaccines have 5G transmitters in them.
Re: (Score:2)
Before the internet, if you wanted to know something you had two choices. You could ask your aunt who always seemed to have the answers, or you could go to the library.
Turns out your aunt just confidently told you whatever nonsense she had been told as a child, and actually knew very little. The library was better, but many books were either out of date or contained inaccuracies, deliberate or accidental.
Huh. Makes you wonder how many of those books were written by Aunts.
Now we have an even more dangerous situation where people think they can get the truth from the internet, so they "do their own research" (google search and reading some Facebook posts by their aunt) and end up believing that COVID vaccines have 5G transmitters in them.
The internet has always been a cesspool of information and bullshit. Putting a label like "AOL" or "Facebook" on it didn't change that. Trolls have been around since the BBS days.
The real dangerous situation happened when "news" reporters started turning to the internet every time as their "source" to ensure they gathered the clickbait to sell faster than the competitor. After a while, only clickbait mattered. Not truth or facts. Beca
Re: (Score:3)
What you want is a "reference encyclopedia". As a child I got gifted an outdated "Der Grosse Brokhaus" (the first "Post Hitler" edition, they did not publish one during the 3rd Reich) and it does show how incredibly neutral and yet accurate and factual you can describe things.
Re: (Score:3)
Given the era it was from that encyclopaedia very likely describes homosexuality and an illness or criminal act.
Nope to "illness", yes to "criminal act" in some countries I expect. You know, facts not morals.
I bet it has some questionable opinions on female psychology too.
Nope.
Seriously, this only shows _your_ prejudices. You should really have a look at a reference encyclopedia some time. They have higher standards and they do not contain or push moral judgements. One of the reasons there are so few of them. They offend every true believer.
Re: (Score:2)
I'd like to have a better search engine, but the ChatGPT needs to solve the problem of "confidently explaining completely false information" before it can replace a search engine.
Oh yes. ChatGPT better get right on that. After all, today's search engines wouldn't ever contain completely false information. /s
(Do you really need AI to explain why false clickbait is for sale to the highest page rank donor? Ironically enough, I fully expect AI to replicate the seven deadly sins. Greed being first.)
Re: (Score:2)
ChatGPT is not a search engine, or even a question answering system. It is, in a strict technical sense, properly called a "bullshit generator".
In 1986 analytical philosopher Harry G. Frankfurt wrote an essay On Bullshit [wikipedia.org] that established a precise explanation of the term and discussed its implications at length. The difference between bullshit and lying is that while lying is a deliberate distortion of what is know to be true, in a real sense the liar knows the truth and cares about it, bullshit does not
Re: (Score:3)
Re: (Score:3)
Because it's way shorter. That's sort of like "why do you need to read the dust-jacket summary of the book when you could just read the book".
I feel like people are forgetting that the Internet is already full of crazy bullshit and if you want to verify the answer you have to follow citations. And most people don't most of the time but it's useful that they're there for those times when it's very important.
This debate reminds me of the "why would anybody ever use wikipedia when anybody can edit it" argume
Re: (Score:2)
Because it's way shorter. That's sort of like "why do you need to read the dust-jacket summary of the book when you could just read the book".
It's really not. For example, I asked Google "How old is Nintendo?" and it gave me this answer [imgur.com]. Two words, correct and brief, no AI needed. I don't need to go anywhere to verify that it's correct, there's a high probability that it is correct.
Re: (Score:2)
Re: (Score:2)
This debate reminds me of the "why would anybody ever use wikipedia when anybody can edit it" argument./quote. Wikipedia has a significantly higher chance of being correct than chatGPT.
Re: (Score:2)
I love it when people explain to me that Wikipedia must be wrong because "anybody can edit it!".
Re: (Score:3, Insightful)
Yep, funny. People without even a base understanding what "review" and "moderation" means, but so full of themselves that they think they understand everything. Same level as flat-earthers and other completely disconnected fuckups.
Come to think of it, ChatAI also can spout the most extreme nonsense while being completely self-assured it is right (or at least pretending to be). Is there maybe a connection between the two?
Re: (Score:2)
I feel like people are forgetting that the Internet is already full of crazy bullshit
And that is the point. Public, non-verified sources are something you need to verify yourself. Of course, the usual disciples of the great tech God "Internet" and its Archangel "AI" feel that there is no need for any fact-checking and that it may be outright blasphemous, but true believers hardly ever have truth and quite often do massive damage.
Re: (Score:2)
I wonder who even moderates ChatGPT and by what parameters and how it gathers information and incorporates it into. How hard would it be to inject falsehoods, slander or misinfo if you knew where it was scraping info from? The more esoteric / specific the topic, the more effective the lie since it's unlikely to be diluted by contradictory information. The same could be said of Wikipedia, but at least it has a lot motivated moderators trying to keep it on track.
Re: (Score:3)
Bings approach just feels like a smart search engine. A bat conversation compared to ChatGPT. ChatGPT makes shit up, but that seems more authentic ...
No... it won't... (Score:2)
All what is going to happen is that eventually, ChatGPT functionality will be built into the search engine, be it Bing's, Google's, or whomever's.
In fact, this is how Google got to be the leader of the pack back in the days of Yahoo, Altavista, HotBot, and so on. Google's search results were generally relevant, while one had to filter a ton of stuff with the others. All AI advances are going to do is just ass more features to the Google prompt.
Re: (Score:2)
Vested interest speaking here (Score:2)
Bill Gates still owns $12.6 billion worth of stock in the company he founded. Microsoft in turn invested billions into OpenAI. Of course he hopes it will displace Google Search and Amazon.
Re: (Score:2)
Re: (Score:2)
Perhaps you haven't noticed, but Bing now incorporates technology from OpenAI. So of course he's not forecasting the end of Bing. He's hoping this will be the enhancement in Bing that will finally get people to leave Google.
The Road Ahead (Score:5, Insightful)
Re: (Score:3)
In his defense.. once he realized what it was, he converted his whole company faster than just about any well established company of the time.
Re:The Road Ahead (Score:5, Insightful)
In his defense.. once he realized what it was, he converted his whole company faster than just about any well established company of the time.
Which, if you think about it, is basically Bill's business strategy from the start. Neither DOS nor Windows were the first operating systems of their types, and neither were the best by any stretch. But he didn't need the best, he just needed something good enough to allow him to entrench Microsoft within the PC industry. When the internet threatened that position, he simply applied the same business strategy to entrench Microsoft in the internet.
He's never been a visionary, although he spends a lot of time trying to pretend he is.
Re:The Road Ahead (Score:5, Insightful)
Re:The Road Ahead (Score:4, Interesting)
Gates obtained that OS from a local hobbyist who had named it QDOS ...
Your post is mostly correct, but QDOS (originally called that as a joke, but by then renamed 86-DOS) was bought, not from a hobbyist, but from a company called Seattle Computer Products, where it had been written by Tim Paterson. MS also hired Paterson to port it to the IBM PC, when it became PC-DOS.
If IBM had a clue they could have bought it directly from SCP, but they did not take personal computers very seriously at the time so didn't do any research into what was available. They approached MS (after getting nowhere with Digital Research) because Gates' mother was a friend of one of the IBM directors.
Re: (Score:2)
Re: (Score:2)
another statement as interesting as his book I never opened.
You should read it to enjoy the bullshit. The style is like one of those popular science books, and is full of the Bleedin' Obvious and self-congratulation.
Of course there soon was a second edition after the first was panned by the critics, with more about the internet. The second edition should have been called The Road Ahead after a U-Turn.
How long until you can be sure data is real? (Score:4, Insightful)
It's truly impressive what AI can do in many cases now.
But to actually be a personal assistant and replace search - that means it would have to be 100% reliable.
Right now you can't be sure at all some of the things the generative AI chat engines are telling you are even real. It will gladly make up facts wholesale, because it has no comprehension what is real and what is not.
dystopia ahead (Score:2, Flamebait)
This yet-to-be developed AI assistant will be able to understand a person's needs and habits
That would be swell if your personal assistant ran locally and kept what you told it to itself. Unfortunately, what will happen is that it will report everything you say and do to its corporate master.
If you thought Big Data was aggressively invasive today, you ain't seen nothing yet. I want none of that future. Whoever comes up with the first personal assistant can shove it where the sun don't shine.
Also, side note: isn't Bill Gates supposed to have retired and busy himself spending his ill-gotten billions
Re: (Score:2)
Also, side note: isn't Bill Gates supposed to have retired and busy himself spending his ill-gotten billions on philantropic projects to buy himself a conscience?
Thankyou!
This is the thing that pisses me off even more about the Bill Gates vax microchip morons, they discredit being anti Bill Gates to a quite large extent. Fuck's sake the man's a crook and burnishing his reputation with the money. No stupid conspiracies needed.
Re: (Score:2)
Indeed. Well said.
Re: (Score:2)
Indeed. Runs locally, is secured against data-theft, goes into my personal backup and can be fully configured by me. Do not see that happening anytime soon.
As to BG, he is just trying to do some more virtue signalling and pretending he understands technology. Obviously he never did.
Re: (Score:2)
They promised "personalized" results (pinky swear) as an excuse to datamine a decade ago to turn what should have been one of the crowning achievements of mankind to little more than a Big Brother Skinner box with ads. Idiocracy was underestimating massively.
And with this next iteration of the web, maybe I can have search results equivalent to what I had in 2007.
Yeah, progress.
Re: (Score:2)
TBH, I want dumber search engines (Score:5, Insightful)
I'm tired of Amazon showing me shit I didn't search for. If I search for a specific model of ThinkPad, either show me that one or nothing if there's nobody selling it. Don't try to outsmart me by assuming I'd like to see a bunch of similar laptops, sometimes "no results found" is the correct answer.
Google does the same thing, by assuming you meant to search for something similar to your input query. No, Google, if that's what I wanted, then that's what I would've typed into the damn search bar.
AI is just going to make playing the game of getting a search engine to actually spit out what you're searching for even more frustrating.
Re: (Score:3)
I have no problems with similar products being listed after the exact match for generic items. Ideally with a divider with exact matches/'no matches found' above and similar below. Valid examples for a laptop would be the newer model of the model searched for or other models with a similar spec and price. What infuriates me is when searching for something like a #8x3" screw and getting all different sizes. Or searching for a replacement part for a specific model of something (such as a laptop battery an
Re:TBH, I want dumber search engines (Score:4, Funny)
I think you and I should have a beer and yell at clouds together.
Couldn't agree more...
Re: (Score:2)
I'm tired of Amazon showing me shit I didn't search for.
For the life of me, I cannot understand how Amazon became so successful with such a terrible website. It is incredibly hard to find something when you know exactly what you want, because it's trying so hard to spam you with adverts. Even the general interface is just gaudy and bloated. It's like the anti-thesis of Apple style design.
Yet Bezos has multiple giant super yachts and rocket company so it's obviously working for him.
Re:TBH, I want dumber search engines (Score:4, Interesting)
The Amazon web site used to be a lot better at showing the items users were looking for. Then Amazon decided to monetize search results directly, and the quality went to hell.
Still, it's hard to imagine that a personal agent will replace Amazon. Their primary advantage now is their huge logistics base. Smaller stores are not going to be able to fulfill orders from hundreds of warehouses around the US without charging more than Amazon does.
Re: (Score:2)
Re: (Score:2)
Amazon pretty much morphed into AliExpress. Except with a domestic markup. Shipping times can be overnight or three weeks, who knows.
Re: (Score:2)
It's even worse if you do not have exact brand in mind.
I recently needed a very small managed (VLAN-capable) switch. There doesn't seem to be a simple way of telling Amazon "list all your network switches that have VLAN capability, sorted by price, starting with cheapest".
Eventually I just grabbed the listing of all their switches and found out that Zyxel and TP-Links offerings were perfect for my purpose (buried somewhere in page 4 or 5). But even their product category "switches" for some reason has PoE i
Re: (Score:3)
Re: (Score:2)
Google does the same thing, by assuming you meant to search for something similar to your input query. No, Google, if that's what I wanted, then that's what I would've typed into the damn search bar.
In one hand I certainly recognize your complaint and have endured it before as well.
On the other hand, there is a significant portion of the search engine user community (which given societal norms is now ages 8 to 80) that often reaches for a search engine because they barely know what to search for. Or even how to spell it correctly. A lot of them are using little more than voice to search.
In those cases (which there are likely plenty), the search engine is assisting. Sometimes greatly.
- Jeeves & Wooster - (Score:3)
TV series starring Fry & Laurie, 1990s. Jeeves was Wooster's man servant who expedited whatever needed doing. Wooster was an irresponsible young rich person who was helpless without Jeeves.
Such servants were fairly common in literature, micromanaging the lives of their 'masters' and covering up their mistakes.
This is what I suspect Gates is referring to: a digital guide to getting thru life successfully despite your instincts that lead you down the wrong path.
But if it is true, it may only be available to the wealthy.
Re: (Score:2)
The wealthy can already afford to hire a Jeeves...
P. G. Wodehouse (Score:3)
Although Hugh Laurie and Steven Frye produced a classic series based on his stories, P. G. Wodehouse is the author that created the Jeeves character.
https://en.wikipedia.org/wiki/P._G._Wodehouse/ [wikipedia.org]
Still salty (Score:2)
Gates is still salty about Google about losing the phone market. Tough shit Bill, nobody wants windows on a phone.
Re:Still salty (Score:4, Insightful)
Nobody wants Windows on anything. We just put up with it on PCs because we have to, but without that stranglehold on phones, there was no way in hell people would put up with it.
Re: (Score:2)
Meh, having been forced to finally give up my BB10 phone, I'd happily take a Windows phone over the mess that is iOS and Android.
Re: (Score:2)
The ironic part is the shitty Windows 8 live tile interface worked really well on a phone. Walmart had $100 phones discounted to $19.95 so I bought one. It actually ran well and was responsive for being a low end phone. It lacked most of the apps people use but I only use a handful even on my iPhone.
This insight is from... (Score:2, Insightful)
This insight is from the same guy that told us 640K was enough. *head scratch*
JoshK.
He's burying the lede. (Score:5, Interesting)
Re:He's burying the lede. (Score:4, Insightful)
The implications of "personalized search results" run way wider and deeper than just that. It starts with the question how it supposedly knows what my "personal" results should be. That information right there is already way, way more than anyone should be comfortable with sharing with some corporation, twice so a corporation that makes its money by selling your personal information (which, at this point, it's probably save to say is all of them). Then the filter bubble feedback loop you describe, which will be even more potent than what we already experience with the manual, "human driven" version we have today, where people surround themselves with like-minded people who keep telling them the same things, which are also things they want to hear and thus are very willing to accept as true.
Imagine this now 24/7 from an AI that is programmed to keep reinforcing your world view, constantly. There isn't even the possibility of a slight deviation because the humans you now interact with might have at least slightly deviating world views, with some more radical than yours that may make you wonder whether they are sane and some less radical than yours that may keep you from going off the deep end yourself, but this AI will keep feeding you the feedback of your own position.
Everyone will live in his perfect little world with no deviation, and everyone will know for a FACT that they are RIGHT. And everyone else on the highway is going the wrong direction.
Discussions as we know them will be over. What we'll have is the internet version of the Jerry Springer show, with lots of yelling and name calling, just with less hair pulling. If only for the lack of the ability to do so.
You just described Facebook. (Score:2)
Where people have created their own echo chambers both by willfully blocking shit they don't agree with and via algorithm.
I'm sure this isn't the only one but it's surely the biggest and among the most dangerous, maybe second to Twitter, for misinformation such as the antivax lunatics drinking their own piss as a panacea.
Re: (Score:2)
Today we call it “doing your own research”
Re: (Score:2)
Heard that once before... (Score:2)
Re:Heard that once before... (Score:4, Informative)
It is actually one of the longest-running empty (so far) claims that the AI community likes to make. The personal "slave" they have been promising for as long as there is AI research. Never panned out and will not pan out this time either. The tech would need AGI to really do that and AGI is not even on the distant horizon.
Re: (Score:2)
A useful personal computerized agent is only five years away from successful commercialization, and has been for the last 40 years, to coin a phrase.
Re: (Score:2)
Indeed. I expect it will be 5 years away for the next 40 years as well.
What is possible with current levels of agency and "insight" is basically reminding you of dates ("Hey Siri, set a timer for 15 minutes..."), but that is something my email-system and my kitchen-timer, respectively my cooking-field and stove, already can do.
Re: (Score:2)
The performance in providing natural language responses, some of them not even nonsensical or false, to natural language prompts has improved; but only at the cost of moving to hardware and training sets that well beyond the reach of almost anyone except as-a-service-potentially-with-the-vendor's-motives-at-heart.
Microsoft has floated a roughly 10x price increase [arstechnica.com] merely for the feature of not having
Re: (Score:2)
I'm not naïve enough to imagine that some ugly FOSS self-hosted thing where you really need to be up on your boolean operators and search-fu will have a chance at the consumer level; but this round suggests that adding even a fairly limited and fallible EZ-natural-language interaction option will more or less guarantee that the agent won't be your agent; even if it's willing to be friendly and helpful in an attempt to guide you toward its operator's preferences for you.
The beauty of FOSS is that it does not need to be successful at the consumer level. It is quite enough if it is successful at the expert level. If wide consumer adoption would be needed for FOSS success, we would not have a Linux kernel (or the xBSD kernels), any Linux/xBSD distros, LibreOffice, GCC, GIMP, EMACS, LaTeX, Python, Perl, etc...
Mere "consumers" will always be lusers, and they will always be an exploitation target for most businesses. Hence I fully agree that operators will try to push their own
So BG does not read books for entertainment? (Score:2)
Figures. Never saw him as somebody cultivated.
Censorship may kill it before (Score:3)
In response to multiple complaints that we received under the US Digital Millennium Copyright Act, we have removed 25 results from this page. If you wish, you may read the DMCA complaints that caused the removals at LumenDatabase.org: Complaint, Complaint, Complaint, Complaint, Complaint, Complaint, Complaint, Complaint, Complaint.
He's smoking crack (Score:3)
Re: (Score:2)
Probably he was thinking about aggregators. Google is a web site aggregator, Amazon is a shopping aggregator. If you can have an IA agent look for what you want, you don't need an aggregator.
But... There's this pesky little problem with the Internet being rather big. If I need a datacenter the size of Google's in order to run the IA and crawl the net, is it any good? On the other hand, if the IA uses aggregators to save resources, how is this different from what we have now?
Of course, there's this third p
Re: (Score:2)
AI can weed through the trash results of Amazon’s search and find what you’re looking for.
Re: (Score:2)
Says who?
If AI can do that then why can't Amazon themselves just do that for me on the first place?
Re: (Score:3)
If AI can do that then why can't Amazon themselves just do that for me on the first place?
Please, please, please think for at least one half of one second before you ask questions with such painfully. obvious. answers.
It's the same reason retail outlets keep shuffling everything around so you can't find it, and orient aisles to make you walk further through the store than you should have to in order to get where you're going. It exposes you to more items that you might buy. Even Netflix does it now! They not only move the stuff you've watched recently to the bottom of your list, but they are con
Ever tried to buy batteries off of Amazon? (Score:4)
Whilst I appreciate Amazon for what it is, it still can't help but show me every other battery type that I don't need, and makes the entire online buying experience a painful one.
Re:Ever tried to buy batteries off of Amazon? (Score:4, Interesting)
As bad as Google is these days, it's a better way to find stuff on Amazon than Amazon's search. That's super-duper sad.
The way a personal software agent could kill Amazon is by handling all your purchases from disparate sites for you. It's very rare that Amazon has the cheapest price on anything, but it is one stop for getting pretty much everything. It's just convenient to put all that stuff in your cart, click one button, and it's ordered.
Re: (Score:2)
Even when I find a slightly cheap price elsewhere I'll still buy through Amazon because their return policy is free and super easy. They even sent UPS to my house the other day for free to pick up my return.
Most of the time when I see a lower price elsewhere I have to pay for return shipping and there's a 15% restocking fee. It only takes one of those to overwhelm any savings I might get from lower prices elsewhere.
Does Bill know what amazon does? (Score:2)
Amazon's search value has always been pretty awful for anything without an ISBN, and they appear to be willingly diluting it to get more sketchy 3rd parties onboard and buying logistics services from them; but the actual logistics operation isn't messing around.
If there is, indeed, going to be a battle of 'the personal agent', apotheosis of Clippy, then I assume that the winner will be pretty well plac
equivalence (Score:2)
could != will
also could != should
..ok then... (Score:2)
.. so I ask AwesomeBillGPT a question and it comes back with an answer and 500 citations of its sources. I look up the citations and a third of them 404, a third are behind paywalls and the remainder are products of another GPT which each have 500 citations. I look up these citations and a third 404, a third are behind paywalls, etc.
Amazon? (Score:2)
Re: (Score:2)
"Ok, here's a poorly written paragraph of text about baseball gloves"
Amazon Alexa (Score:2)
already tried this. How did that work out?
Apple's Knowledge Navigator (back to the future) (Score:2)
Good morning Mister Gates! (Score:2)
Big Tech already has... (Score:2)
Be careful what you wish for (Score:2)
Just think how fast a "personal digital agent" could have helped Melinda Gates find the link between her ex and Jeffrey Epstein.
Of course, it's probably easier to play "Six Degrees of Jeffrey Epstein" than the Kevin Bacon version of the game, so maybe this wouldn't offer much of a challenge to our fledgling Skynet.
As Predicted .... (Score:2)
This is what was predicted in 1990 ....
By Douglas Adams in Hyperland : https://en.wikipedia.org/wiki/... [wikipedia.org]