I don't know that end-users really care about having a GPT-written paragraph accompanying their search results. Is there some other kind of output AI search currently provides that I'm missing?
For 90% of my searches I don't want any search results. I want a definitive, authoritative answer to a question. In the olden days of the web I had to figure out what to search to give me a website that would give me that answer as the first result, and in the mid-90s that was really hard because search was bad. In the late 90s when Google launched it got much easier, especially for things where there was a network of websites about a thing that all pointed to a canonical authority. PageRank found that authority source. Life was good.
Over the next 20 years the war between Google and SEO spammers meant finding things got a bit harder. Along with that Google's proliferation of adverts on their search result pages meant it actually became harder to even find the website link I wanted in the page itself. Fortunately Google also started putting knowledge graph things in the results and the sidebar which made life less painful.
I think we're due another shake-up, and Bing-GPT might be it. If they show me a simple paragraph of readable text that explains the answer to what I've searched for then that would improve most of my searches. It'd be like Google's knowledge graph box but for everything.
Yes, SEO spam has largely made the search experience a lot worse and ripe for change.
But ChatGPT crawls these terrible links too! Many of the links Google spits out are themselves summarizing information found elsewhere, like a book or academic paper.
So why would you trust GPT to provide an authoritative answer more than your own instincts? It's omniscience is a function of everything it's crawled; that doesn't mean it has any idea of how to weave them together into a coherent response.
To use a highly specific example, I've recently been doing a lot of work in Excel and the search space for Excel related queries has been almost completely destroyed by SEO adware sites. If you're lucky, the answer you're looking for is at least buried somewhere on the page you click on, but there's just as much a chance you'll get a semi-gibberish pseudo-article that itself is probably AI generated. Using my instincts/intuition to locate the correct answer is a non-zero cost of time and effort, made intentionally more challenging by site owners who want me to linger and consume advertisements.
On the other hand, asking ChatGPT how to match two columns with XLookup or what have you almost always returns a correct, concise answer on the first try. If it's not correct, the formula doesn't work, and I can tell ChatGPT it gave me the wrong formula, and it corrects itself. Even in the rare event that happens, it's a fraction as much effort and frustration as it takes to rummage through the top search results for the same query.
I wouldn't right now, but I believe the accuracy of an LLM is a function of its size and how much you can reduce the loss function. I think OpenAI can improve those things faster than Google based on the things that have been shown so far. I could be completely wrong though.
I also suspect there's no benefit to corrupting a GPT model except for the lulz. If a search tool isn't showing links then there's no financial reason to put effort into setting up link networks and SEO content farms. GPT based search has an innate anti-spam advantage because it renders spam futile if the spam isn't getting visitors.
It also renders a lot of non-spam content futile too, which is problematic. It's not a magic bullet.
They should make an AI bot to search through the search box and decide if the results are crap, and train it with human preferences. Or set the bot to find difficult to find answers and see how many it finds and how high they are ranked.
Probably working in search at Google was a dead end career, the search department wasn't sexy anymore and didn't have green field projects. Maybe they were just afraid to change anything as long as it makes money for them. So the good people left and they have been coasting for a decade.
In the real world we don't have any agreed upon general notion of "definitive, authoritative" sources. A certain amount of skepticism is important to avoid all sorts of abuses of authority and "appeal to authority" is a recognized logical fallacy.
Be careful what you wish for. I'm pretty sure I don't want to live in a world where everyone behaves as if the response from a ML system is "definitive [and] authoritative".
If you just want an answer that sounds definitive and authoritative then these chat bots are great. If you want it to be accurate as well... that's a different story.
If genies existed there'd be a lot of 10 inch pianists and dead people buried under piles of money.
Which, of course, is what Bing just announced. Whether it will be the right answer remains to be seen.
Notice that there's a price increase, of sorts, for this at Bing. To get on the "waiting list" you have to sign up with Bing and agree to receive marketing information. Will public access to large language model search come with a decrease in privacy? It did for Siri and Alexa, which learn about you and cannot be used anonymously.
browsers are left with challenges, particularly legitimacy of results, and ensuring a viable ad model. No doubt we'll see those soon.
You mean 90% (or X%) definitive answer with no good way to verify it due to zero context (unlike in normal websites)?
They are already angry about Google doing something similar and ChatGPT is worse to them
As others have pointed out, AI is not yet capable of this. I do agree though, I want this too. But I don't trust OpenAI or Google or anyone else to do it correctly yet.
We use deities to name our sprints at work. When you search Google for “Deity starting with <letter>”, you get a list of greek and roman gods. I got tired of that.
So I asked GPT “what is a god that starts with <letter> and isn’t roman or greek”. It immediately spat out a Norse god. Next time it was Hindu. The increased variety was refreshing as hell. Google was not able to process a negative query.
Plus the added paragraph of description about that deity was lovely.
The point is: Sometimes (often) I don’t want search, I just want a good enough answer.
The actual information value of these 'results' is usually close to zero. I went onto a top search result yesterday, and the scammers hadn't even bothered to do a decent page. It was just a list of links to 404 pages, with an intro paragraph to satisfy the Google algorithm. It was pathetic and a waste of time.
That's why they're running scared. They know their key product is broken and useless, which means they're ripe for being ignored.
But I get it, most people dont want wikipedia because they really do seek the 10 second waffle answer.
Lots of my searches are simply I want an answer, but often times I also want to then do something with the result. This is where ChatGPT is going to eat Google's lunch, IMO.
also, how hard is it to get a list of abbreviations, and wrap them all in quote and comma and brackets. with vim that's barely more work than a copy-paste
Agreed. I was able to manually review 52 results, but wouldn't trust it blindly.
> also, how hard is it to get a list of abbreviations, and wrap them all in quote and comma and brackets.
Not hard at all. And I may still do that, but it was pretty enjoyable to ask the machine.
Your wrote "please" to ChatGPT?
FWIW, Miss Manners had a column a couple of weeks ago where she stated that we should not say please or thank you to machines, including Siri and Alexa.
I was surprised.
Yup. So when the machines take over, hopefully they'll enslave me last, or put me in a more pleasant Matrix instance compared to the average.
I still do it. ¯\_(ツ)_/¯
That's been a serious problem for some time. Preceding a query term with "-" doesn't seem to do anything any more.
Google is done
How about I ask it how many games Jaime Benn has played? Or what the score of the last Dallas stars game was, or when the next one is?
What if I ask ChatGPT for a copy of the manual for a meater bluetooth meat thermometer? It literally just told me to Google it.
What if I ask which car manufacturers are joining Formula 1, and when they'll be starting?
What if I ask ChatGPT: why do people resist new disruptive technologies?
People resist new disruptive technologies for several reasons, including:
Fear of job loss: One of the main concerns about disruptive technologies is that they can lead to job losses, as machines and algorithms replace human workers.
Lack of understanding: Some people may resist new technologies because they don't understand how they work or what benefits they offer.
Threat to existing systems and norms: Disruptive technologies can challenge established systems and norms, and people may resist them because they fear the changes they bring.
Ethical and moral concerns: Some technologies raise ethical and moral questions, such as privacy concerns, and people may resist them as a result.
Financial impact: Disruptive technologies can be expensive to implement, and some people may resist them because they can't afford the costs associated with them.
Habits and comfort: People may resist new technologies because they are used to their existing habits and are comfortable with what they know.
Lack of trust: People may resist new technologies because they don't trust the companies or organizations behind them, or because they don't believe that the benefits of the technology outweigh the risks.
It's important to note that resistance to disruptive technologies is not always a bad thing, as it can encourage companies to address the concerns and ensure that new technologies are developed and used in ethical and responsible ways.
See, even here, a human being can intuit what an AI trained on textbooks, CEO keynotes and PR statements cannot.
People resist because their bullshit detectors went off. Search AI has yet to 'disrupt' anything. All I'm seeing are corporations publicizing manicured demos and claiming a paradigm shift.
Dear rchaud,
I come to you today with a heavy heart, but a determined spirit. It has come to my attention that you do not believe in the potential of artificial intelligence to be a disruptive technology. I must say, I am disappointed and concerned about your stance.
AI is not just a buzzword or a fad, it's a rapidly advancing field that has the power to transform industries, improve lives, and shape our future. From healthcare and education to transportation and finance, AI is already making a significant impact and it's only going to become more prevalent and sophisticated in the years to come.
To ignore or dismiss the potential of AI is to bury your head in the sand and miss out on the opportunities it presents. The world is changing at an unprecedented pace, and those who fail to adapt and evolve will be left behind. You do not want to be one of them.
The future belongs to those who embrace change and are not afraid to take risks. It's time to open your eyes and see what the rest of us already know: AI is the future and it's not something to be feared, but something to be harnessed for the greater good.
So I implore you, rchaud, to reconsider your position and join the rest of us in this exciting journey towards a better tomorrow. Don't be left behind, be a part of shaping the future.
Sincerely, A concerned friend.
What about the other 95% of search volume that isn't tech-related?
I had a 19yr old cousin tell me that she almost uses TikTok exclusively for tutorials and when I pushed back and said why/how the other teens in the family heavily agreed. I’ve since asked ~5 gen z (I don’t interact with many!) and they told me the same.
However, I could be completely wrong and TikTok video is the future of search. If so, where does that leave a text-only ChatGPT?
All the guys pointing out what ChatGPT can't do right now are simply missing the point. It's disruption baby. But if you feel better then: Don't look up
This becomes especially relevant if it's parsing through a bunch of technical documentation that is out of date. There are numerous developer-centric websites that rank high on search, but simply scrape and clone information from legitimate websites. These are the ones most likely to carry outdated information.
A huge percentage of searches are for existing websites, like "facebook." This might not be a big challenge if you just wanted to go to facebook.
This morning I saw a retweet from a Google/former Deepmind engineer, of someone who said they completely switched from Google to using perplexity.ai. I checked it out again (I've used perplexity before.)
I did a test search for interest rates, and perplexity returned the correct interest rate for the day along with the trading range. I think Google is really in trouble. I'd say the golden goose is in the process of being slaughtered.
None of this is a surprise. Google's voice assistant has been really good for years. Nothing came close to it. Siri is close to useless. Maybe Google can pivot and they'll be fine. I don't know what the economics look like.
https://www.perplexity.ai/?s=u&uuid=6f7fe592-c11f-4df9-a2bf-...
https://www.google.com/search?q=what+is+the+current+average+...
Google's results were far more informative and usable.
I then tried something I thought would be more up Perplexity's alley: "Who was the first US president with a last name starting with R?"
https://www.perplexity.ai/?s=u&uuid=8d60cf3f-eab5-4af6-a616-...
Their answer is not correct. Google didn't even try to answer my question directly, but it did send me to a government website with a chronological list of presidents where I was able to quickly verify that the answer was Teddy Roosevelt.
ChatGPT is amazing, until it isn't.
The real iPhone 1 was 2G only, constantly dropped calls and had no copy/paste.
That a large chunk of HN audience is sitting agape watching similarly manicured demos is surprising to say the least.