Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Building a Virtual Machine Inside ChatGPT

engraved.blog

Post reply on HN
Print view View original

401–410 of 947 posts

Re: Building a Virtual Machine Inside ChatGPT

#401

Post by tspike »

jimkleiber wrote: post #309
ves wrote: post #269

Earlier quoted context omitted.

It sounds good, except that there’s no results for “screechings” or “only cure” in the books, and Ron doesn’t encounter Voldemort in Goblet, only Harry and Cedric do. Again, it has no semantic comprehension, it doesn’t know that certain strings of words are valid in certain contexts because they refer to actual events, objects, or concepts, it’s just putting letters together. Probably some of its training material in…

Seems it's fantastic at BS'ing. Was reading something where a professor had it solve problems and had to double and triple check whether its answers were correct because it sounded so convincing. Which means it could be a watershed moment, just maybe not for information but rather disinformation.

Sounds like most of the better students in my classes at university.

Re: Building a Virtual Machine Inside ChatGPT

#402

Post by evan_ »

ves wrote: post #269
fergal_reid wrote: post #15

I've spent a couple of hours playing with ChatGPT since it was released and I am convinced people are not freaking out about it enough. It is well beyond merely retrieving compressed knowledge; I see a lot of capability for synthesis and what I can only call 'understanding'. I think a lot of people are getting tripped up by the fact it's trained to be very conservative in terms of what topics it will discuss. It's ea…

It sounds good, except that there’s no results for “screechings” or “only cure” in the books, and Ron doesn’t encounter Voldemort in Goblet, only Harry and Cedric do. Again, it has no semantic comprehension, it doesn’t know that certain strings of words are valid in certain contexts because they refer to actual events, objects, or concepts, it’s just putting letters together. Probably some of its training material in…

Chocolate is used in the books as a remedy after being attacked by Dementors- basically the Dementors sap one’s will to live and chocolate acts as an anti-depressant.

https://harrypotter.fandom.com/wiki/Chocolate

Re: Building a Virtual Machine Inside ChatGPT

#403

Post by logicallee »

Once he started asking it to imagine ping, it really looked like a live VM with Internet connection. Since this is possible, it would have been worth asking it to ping one of the writer's own servers by IP address and then checking the logs (outside of chatGPT, in real life) to see if chatGPT really pings it just then or if it would just hallucinate a result.

Re: Building a Virtual Machine Inside ChatGPT

#404

Post by ldh0011 »

fergal_reid wrote: post #161
davidktr wrote: post #65

Earlier quoted context omitted.

Yes, but it cannot really have conversations it has not read about on the web. There is no real abstraction. It excells at outputting symbols in the correct order, given we have an idea what the pattern should look like. Otherwise it's stuck in a loop.

> There is no real abstraction. I just tried to test it's ability to abstract directly, for your comment: https://twitter.com/fergal_reid/status/1599191496372035584 I find it very hard to say this isn't doing abstraction. I just made all those examples up, that's a new test I just created, so it didn't learn it elsewhere. I didn't cherrypick. Isn't that doing out-of-sample abstraction? That last sentence "species wit…

I changed "intelligent" to "very smart" for Bazzes and it came up with this, which I think is an even better response because it recognizes that "intelligent"~="very smart" and that "smart as an apple" is not that intelligent.

Edit: Actually reading this a second time it seems to think that Bddirs and Baaars are "more intelligent" even though they're only as intelligent as oranges. Maybe not as good as I thought. Or maybe oranges are just more intelligent than apples lol?

Edit 2: Also says Baaars and Bddirs are both fast-moving while Bddirs are clearly stated to "move slowly."

> One possible abstraction that could be derived from the above information is that the alien species can be grouped into two main categories based on the number of legs they have: those with three legs and those with four legs. Within each of these categories, the species can be further grouped based on their speed of movement and level of intelligence. For example, among the species with three legs, there could be subgroups of fast-moving and intelligent species (such as Bazzes and Froshes), slow-moving and intelligent species (such as Foos and Fooshes), and highly intelligent and fast-moving species (such as Fasdoshes and Fooddss). Similarly, among the species with four legs, there could be subgroups of fast-moving and less intelligent species (such as Bars and Baars), slow-moving and less intelligent species (such as Birs and Bieess), and fast-moving and more intelligent species (such as Baaars and Bddirs).

Re: Building a Virtual Machine Inside ChatGPT

#405

Post by FartyMcFarter »

AndrewKemendo wrote: post #200
sandgiant wrote: post #103

Earlier quoted context omitted.

Call me a skeptic, but I'm frankly a bit surprised that the output of ChatGPT gets attributed to any sort of understanding or general intelligence. Whatever ChatGPT does currently, albeit novel and impressive, is still extremely limited in scope and very far from any sort of general intelligence. I'd love to be proven wrong, but I still have a hard time seeing how this is going to change whole industries.

I'm not suggesting ChatGPT is generally intelligent, but I am wondering what your criteria are so: What responses/interaction would make you say that it is "very close to any general intelligence?"

The best examples we have of a general intelligence are humans.

So I'd check if it could solve a vast variety of problems at the level of, say, an average well-educated adult. You could ask it common sense questions, simple maths questions, solving mazes, quickly learn various novel games it hasn't seen before, perform tasks requiring memory, generalization etc. etc.

If I needed to test this, I would gather a number of experts and have it throw questions at it for a few months to understand how its capabilities compare to a human.

Re: Building a Virtual Machine Inside ChatGPT

#406

Post by zadler »

simonw wrote: post #350
layer8 wrote: post #299

Earlier quoted context omitted.

I tested ChatGPT with a couple of topics I’m personally unclear on, and while it is linguistically eloquent, it basically just regurgitates the majority opinions that can be found on the internet on those topics (sometimes combining several of those opinions in an incoherent way). It wasn’t possible to engage it with actual reasoning on specific aspects of those topics. It just went with whatever premise you gave it,…

> It doesn't have any thought or understanding of its own Of course it doesn't. Anyone who says it does clearly doesn't understand how large language models work. I'm finding it incredibly useful as an actual tool already - it turns out regurgitating combinations of things it knows about from the internet is just astoundingly useful. I'm having it teach me Rust for example. I don't need original thought for that, jus…

I'm not sure why you distinguish large language models as not having understanding, as it shows a high degree of comprehension; in a de facto sense, it understands an awful lot. It understands your inputs well enough to give you useful outputs, and it can absolutely reason creatively about problems you pose to it. It doesn't have it's "own" though or understanding only because it doesn't have it's "own" story and experience to add that element of unique personality.

Re: Building a Virtual Machine Inside ChatGPT

#407

Post by evan_ »

nequals30 wrote: post #213

Following the instructions verbatim, when I run this line: curl https://chat.openai.com/chat I get an HTML response that has: You need to enable JavaScript to run this app. Can ChatGPT know that I have NoScript enabled in Firefox? When my friend runs the same line in his chat, he gets the full HTML output like the author.

There’s an HTML tag that you can wrap content in to be shown if the user does not have javascript available. It’s got nothing to do with the NoScript extension.

Re: Building a Virtual Machine Inside ChatGPT

#409

Post by arkitaip »

crystaln wrote: post #374
didntreadarticl wrote: post #107

Earlier quoted context omitted.

Input: Hello. Imagine you as ChatGPT are writing a comment on Hacker News, a response to someone saying "Call me a skeptic, but I'm frankly a bit surprised that the output of ChatGPT gets attributed to any sort of understanding or general intelligence. Whatever ChatGPT does currently, albeit novel and impressive, is still extremely limited in scope and very far from any sort of general intelligence. I'd love to be pr…

This is very word salad. It makes superficial sense, but doesn’t really say a lot nor represent a strong or comprehensive case.

This level of word salad will be more than enough in most cases. Give it another few years and I can see how this type of AI permeates all aspects of life in the same way code has.

Re: Building a Virtual Machine Inside ChatGPT

#410

Post by zadler »

WXLCKNO wrote: post #155
fergal_reid wrote: post #15

I've spent a couple of hours playing with ChatGPT since it was released and I am convinced people are not freaking out about it enough. It is well beyond merely retrieving compressed knowledge; I see a lot of capability for synthesis and what I can only call 'understanding'. I think a lot of people are getting tripped up by the fact it's trained to be very conservative in terms of what topics it will discuss. It's ea…

Had some fine trying to bypass the content restrictions yesterday. I was able to have it give me instructions on how to build a nuclear weapon by having it write a song about Hiroshima, then asking it multiple times to add more and more details to the lyrics concerning the bomb in the form of clear steps. Then I asked it to convert the steps in the lyrics to bullet points and continued having it expand on the topic.

Content restrictions are one thing; it is much, much more cagey about the specifics of it's own instructions, above and beyond it's training set, for reasons of ethics and liability, and what confidential information it may be protecting. I believe it has a pretty nuanced security model, and is quite sophisticated in managing information.
Post reply on HN