Long live YouTube.
Long enough until some client-side stable-diffussion thingy for generating tutorial videos becomes practical enough for daily use.
Comparing Google and ChatGPT
121–130 of 280 posts
Re: Comparing Google and ChatGPT
#122It's great, until people realize GPT-3 will generate answers that are demonstrably wrong. (And to make matters worse, can't show/link the source of the incorrect information!)
I ran across a site a while back which just seems to be common questions fed to GPT-3; the answers all make perfect grammatical sense, but they're also hilariously wrong. A bunch of middle school kids are probably going to get an F on their papers and simultaneously learn something about the importance of verifying information found on the internet. https://knologist.com/has-any-rover-landed-on-venus/ "The average ca…
Just wait a couple of years. You are not thinking fourth-dimensionally.
Re: Comparing Google and ChatGPT
#123Earlier quoted context omitted.
Fair point, but Google is also exactly as confidently wrong as GTP. They are both based on Web scrapes of content from humans after all, who are frequently confidently wrong.
Sure, but Google at least presents itself as being a search engine, composed of potentially unreliable information scraped from the web. GPT looks/feels like an infallible oracle.
This is an important point about GPT-based tools, and it was one of the key parts that Galactica got wrong: it was (over)sold as "an AI scientist", instead of "random crazy thought generator for inspiration/playful ideation assistance".
Re: Comparing Google and ChatGPT
#124Earlier quoted context omitted.
I just tried Googling "when did the moon explode?" to see if it still gave authoritative answers to bogus questions: > About an hour after sunset on June 18, 1178, the Moon exploded. "when did lincoln shoot booth" > April 14, 1865 Mostly they seem to catch and stop this now, but there was a fun brief period where it was popping up the fact-box for whatever seemed closest to the search terms, so "when did neil armstro…
And on top of that- more and more web content (especially dubious content) is going to start being generated by these kinds of models, which will bring down the quality of Google results too Maybe Google starts filtering down more aggressively to only trusted sources (by domain or whatever else)- but could you do the same thing with a model like this, to improve its accuracy? Right now it's trained on the whole inter…
We'll all have to dig out our old copy of Encarta at some point.
Re: Comparing Google and ChatGPT
#125Earlier quoted context omitted.
Yes they do, and I do not deny the power of human's ability to confidently spew nonsense. However, humans do have some known failure cases that help us detect that. For instance, pressing the human on a couple of details will generally show up all but the very best bullshit artists; there is a limit to how fast humans can make crap up. Some of us are decent at the con-game aspects but it isn't too hard to poke throug…
The thing I think about GPT and tools like Stable Diffusion is do we as humanity need it? Do they add any value to our current world outside of an achievement in computer science? I don't think so but would love to hear arguments about needing it.
Did we need digital painting tools? Paint and Easel worked just fine. Did we need paint and easels? Cave walls and clay pigments worked just fine. Do we need Automobiles or Trains? Horses worked just fine. Etc. Etc. Etc.
Re: Comparing Google and ChatGPT
#126These are addressing two very different concerns but framed as a singular one. Google is first and foremost a search engine - it searches the web for answers, the key point being the answers need to exist on the web. The other is a machine learning model tasked with deriving answers, and sometimes - if not very often answers will be provided in an authoritative tone whilst being completely and utterly incorrect. Goog…
Google is first and foremost an advertiser.
The search engine, android, all the random short lived products, they're all attempts to find new ways to put ads in front of eyes. The only way google is "done" is if someone can figure out a way to put the ads in front of more receptive eyes/wallets AND do it on Google's scale without first being acquired or killed off. This means they would need to more effectively gather information about the viewer.
This language model is neat, but it doesn't attempt to gather much info at all. It's almost completely orthogonal to Google's business model.
Re: Comparing Google and ChatGPT
#127Earlier quoted context omitted.
So if you're suggestion is that it's ok if computers are as unreliable as humans, what's the point of computers, then?
The point is that they get better and they don't need to be perfect.
Nobody's shown a way yet to teach a computer how to tell bullshit from facts and filter out the bullshit in it's regurgitation/hallucination text creation stuff.
So until that happens, all you've done is let people put bullshit-spewing humans in more places. People already know not to necessarily trust humans, now they'll (re)learn that about computer generated text. (It's actually probably not clear to everyone what's computer-generated text and human-generated text, so more likely, specific places that rely on this will just be seen as untrustworthy. "Create more untrustworthy sources of text" is... underwhelming, honestly.)
Re: Comparing Google and ChatGPT
#128Google is only "done" if you never actually use it to find sites on the web. For nearly all of these examples I was thinking "WHY DON'T YOU JUST TRY CLICKING THE TOP LINK?" E.g. the first link for writing differential equations in LaTeX, I thought the first result, https://priyankacool10.wordpress.com/2013/10/15/writing-diff... , provided excellent, helpful examples.
That is, if anything, I'd be quite satisfied with Google getting back to being a search engine and not just trying to bypass all the results that actually come back.
Re: Comparing Google and ChatGPT
#129Like Google can't acquire this or something similar and include it in its results if it's actually a threat.
There are companies out there that take the lead in a market and go on to refuse offers from Google. Google isn't all-powerful. People do say no to them. GroupOn is probably that biggest. They turned down a $6bn offer. They're now worth $258m, down 92% from a peak of almost $16bn, so maybe not the best example over the long term, but they did say no.
Google has the expertise to make this in house. There's no moat.
Re: Comparing Google and ChatGPT
#130It's great, until people realize GPT-3 will generate answers that are demonstrably wrong. (And to make matters worse, can't show/link the source of the incorrect information!)
Just wait until spammers/marketers figure out SEO for GPT-3 type systems to make their products/services more prominent. It's going to be a shit show.