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

LLM Inevitabilism

tomrenner.com

Post reply on HN
Print view View original

971–980 of 1001 posts

Re: LLM Inevitabilism

#971

Post by yieldcrv »

strange_quark wrote: post #828
lsy wrote: post #63

I think two things can be true simultaneously: 1. LLMs are a new technology and it's hard to put the genie back in the bottle with that. It's difficult to imagine a future where they don't continue to exist in some form, with all the timesaving benefits and social issues that come with them. 2. Almost three years in, companies investing in LLMs have not yet discovered a business model that justifies the massive expen…

> There are many technologies that have seemed inevitable and seen retreats under the lack of commensurate business return (the supersonic jetliner) I think this is a great analogy, not just to the current state of AI, but maybe even computers and the internet in general. Supersonic transports must've seemed amazing, inevitable, and maybe even obvious to anyone alive at the time of their debut. But hiding under that…

slower, no fast option, no smoking in the cabins, less leg room, but with TVs plastered on the back of every chair, sometimes

its actually kind of scary to think of a world where generative AI in the cloud goes away due to costs, in favor of some other lesser chimera version that can't currently be predicted

but good news is that locally run generative AI is still getting better and better with fewer and fewer resources consumed to use

Re: LLM Inevitabilism

#972

Post by aaronbaugher »

jdauriemma wrote: post #732
DanMcInerney wrote: post #712

These articles kill me. The reason LLMs (or next-gen AI architecture) is inevitably going to take over the world in one way or another is simple: recursive self-improvement. 3 years ago they could barely write a coherent poem and today they're performing at at least graduate student level across most tasks. As of today, AI is writing a significant chunk of the code around itself. Once AI crosses that threshold of con…

> they're performing at at least graduate student level across most tasks I strongly disagree with this characterization. I have yet to find an application that can reliably execute this prompt: "Find 90 minutes on my calendar in the next four weeks and book a table at my favorite Thai restaurant for two, outside if available." Forget "graduate-level work," that's stuff I actually want to engage with. What many peopl…

I've found that they struggle with understanding time and dates, and are sometimes weird about numbers. I asked Grok to guess the likelihood of something happening, and it gave me percentages for that day, the next day, the next week, and so on. Good enough. But the next day it was still predicting a 5-10% chance of the thing happening the previous day. I had to explain to it that the percentage for yesterday should now be 0%, since it was in the past.

In another example, I asked it to turn one of its bullet-point answers into a conversational summary that I could turn into an audio file to listen to later. It kicked out something that converted into about 6 minutes of audio, so I asked if it could expand on the details and give me something about 20 minutes. It kicked out a text that made about 7 minutes. So I explained that that was X words and only lasted 7 minutes, so I needed about 3X words. It kicked out about half that but claimed it was giving me 3X words or 20 minutes.

It's the little stuff like that that makes me think that, no matter how useful it might be for some things, it's a long way from being able to just hand it tasks and expect them to be done as reliably as a fairly dim human intern. If an intern kept coming up with half the job I asked for, I'd assume he was being lazy and let him go, but these things are just dumb in certain odd ways.

Re: LLM Inevitabilism

#973

Post by filoeleven »

digianarchist wrote: post #554
Buttons840 wrote: post #475

Earlier quoted context omitted.

We need to put the LLMs inside systems that ensure they can only do correct things. Put an LLM on documentation or man pages. Tell the LLM to output a range of lines, and the system actually looks up those lines and quotes them. The overall effect is that the LLM can do some free-form output, but is expected to provide a citation to support its claims; and the citation can't be hallucinated, since the LLM doesn't gen…

Are models capable of generating citations? Every time I've asked for citations on ChatGPT they either don't exist or are incorrect.

Not sure what the state of it is now, but I saw this video of someone working on a "classifier" that recognizes when the LLM is "trying" to quote something from its training set, provides the actual data from the training, and includes citations.

https://www.youtube.com/watch?v=b2Hp0Jk9d4I

Re: LLM Inevitabilism

#974

Post by jdauriemma »

aaronbaugher wrote: post #972
jdauriemma wrote: post #732

Earlier quoted context omitted.

> they're performing at at least graduate student level across most tasks I strongly disagree with this characterization. I have yet to find an application that can reliably execute this prompt: "Find 90 minutes on my calendar in the next four weeks and book a table at my favorite Thai restaurant for two, outside if available." Forget "graduate-level work," that's stuff I actually want to engage with. What many peopl…

I've found that they struggle with understanding time and dates, and are sometimes weird about numbers. I asked Grok to guess the likelihood of something happening, and it gave me percentages for that day, the next day, the next week, and so on. Good enough. But the next day it was still predicting a 5-10% chance of the thing happening the previous day. I had to explain to it that the percentage for yesterday should…

This is similar to many experiences I've had with LLM tools as well; the more complex and/or multi-step the task, the less reliable they become. This is why I object to the "graduate-level" label that Sam Altman et al. use. It fundamentally misrepresents the skill pyramid that makes a researcher (or any knowledge worker) effective. If a researcher can't reliably manage a to-do list, they can't be left unsupervised with any critical tasks, despite the impressive amount of information they can bring to bear and the efficiency with which they can search the web.

That's fine, I get a lot of value out of AI tooling between ChatGPT, Cursor, Claude+MCP, and even Apple Intelligence. But I have yet to use an agent that has come close to the capabilities that AI optimists claim with any consistency.

Re: LLM Inevitabilism

#975

Post by rafaelmn »

vharish wrote: post #953
rafaelmn wrote: post #24

Earlier quoted context omitted.

If you claimed that AI was inevitable in the 80s and invested, or claimed people would be inevitably moving to VR 10 years ago - you would be shit out of luck. Zuck is still burning billions on it with nothing to show for it and a bad outlook. Even Apple tried it and hilariously missed the demand estimate. The only potential bailout for this tech is AR, but thats still years away from consumer market and widespread a…

What are you on? The only potential is AR? What?!!! The problem is AR is not enough innovation and high cost. That's not the case with AI. All it needs is computing, not some ground breaking new technology.

The only potential for VR is salvaging the investment by pivoting to AR.

And if it's only compute why we are seeing teams with limited compute from china reach SOTA model performance, and teams with a bunch of compute available (Meta) fail ?

Re: LLM Inevitabilism

#976

Post by silexia »

I haven't logged into HN to comment or upvote in a long while as I don't want to play the game of trying to fight to get heard... But this is an excellent point.

Let's gather together and unite to stop AI... I am not concerned about the jobs issue, I am concerned about the extinction of the human species.

Re: LLM Inevitabilism

#977

Post by handoflixue »

madrox wrote: post #894
evantbyrne wrote: post #543

Earlier quoted context omitted.

I'm pretty bearish on the idea that AGI is going to take off anytime soon, but I read a significant amount of theology growing up and I would not describe the popular essays from e.g., LessWrong as religious in nature. I also would not describe them as appearing poorly read. The whole "look they just have a new god!" is a common trope in religious apologetics that is usually just meant to distract from the author's o…

I've read LessWrong very differently from you. The entire thrust of that society is that humanity is going to create the AI god.

They are literally publishing a book called "If you build this, everybody dies" and trying to stop humanity from doing that. I feel like that's an important detail: they're not the ones trying to create the god, they're the ones worried about someone else doing it.

Re: LLM Inevitabilism

#978

Post by habinero »

citizenpaul wrote: post #878
kenjackson wrote: post #845

Earlier quoted context omitted.

I think this is both right and wrong. There was a good book that came out probably 15 years ago about how technology never stops in aggregate, but individual technologies tend to grow quickly and then stall. Airplane jets were one example in the book. The reason why I partially note this as wrong is that even in the 70s people recognized that supersonic travel had real concrete issues with no solution in sight. I don…

Was the problem that supersonic flight was expensive and the amount of customers willing to pay the price was even lower than the number of customers that could even if they wanted to?

Yeah, basically. Nobody wanted to pay $12,000 to be in a plane for three hours when they could pay ~$1200 to be in one for six hours. Plus, they used up a lot of fuel. That made them real vulnerable to oil price spikes.

Contrast that with modern widebody jets, which fly ~300 people plus paid cargo on much more fuel-efficient engines.

Re: LLM Inevitabilism

#979

Post by eschaton »

codebolt wrote: post #943
strange_quark wrote: post #828

Earlier quoted context omitted.

> There are many technologies that have seemed inevitable and seen retreats under the lack of commensurate business return (the supersonic jetliner) I think this is a great analogy, not just to the current state of AI, but maybe even computers and the internet in general. Supersonic transports must've seemed amazing, inevitable, and maybe even obvious to anyone alive at the time of their debut. But hiding under that…

The crucial point is that we simply do not know yet if there is an inherent limitation in the reasoning capabilities of LLMs, and if so whether we are currently near to pushing up against them. It seems clear that American firms are still going to increase the amount of compute by a lot more (with projects like the Stargate factory), so time will tell if that is the only bottleneck to further progress. There might al…

What reasoning capabilities?

Re: LLM Inevitabilism

#980

Post by blks »

nosianu wrote: post #287
blks wrote: post #262

Earlier quoted context omitted.

So how big was the library? If I understood correctly, it was a single file library (with hours worth of documentation)? Or did you go over all files of that library and copy it file by file?

Funny you use something the author of the linked post talks about at the start. This is one of those debate methods. Reframe what was said! I don't remember that the OP claimed that all problems are solved, perfectly. Do you think by showing examples where AI struggles you really show their point to be wrong? I don't see that. I use AI only sparingly, but when I do I too experience saving lots of time. For example, I…

I don’t understand why are you replying to me. I asked a person who wrote a comment about his process. Your comment does not answer my question.
Post reply on HN