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

AI isn’t good enough

skventures.substack.com

Post reply on HN
Print view View original

181–190 of 374 posts

Re: AI isn’t good enough

#181

Post by td118 »

oceanplexian wrote: post #120
alpaca128 wrote: post #114

Earlier quoted context omitted.

> LLMs won't get intelligent Even assuming that is true: LLMs aren't all that exists in AI research and just like LLMs are amazing in terms of language it's possible similar breakthroughs could be made in more abstracted areas that could use LLMs for IO. If you think ChatGPT is nice, wait for ChatGPT as frontend for another AI that doesn't have to spend a single CPU cycle on language.

The next AI wave hasn't even started. Imagine an LLM the size of GPT-4 but it's trained on nothing but gene sequence completion. All the models being used in academia are basically toys, none of those guys are running hardware at a scale that can even remotely touch Azure, Meta, etc, and right now there is a massive global shortage of GPU compute that's eventually going to clear up. We know models get A LOT better wh…

[dead]

Re: AI isn’t good enough

#182

Post by gorbypark »

dontupvoteme wrote: post #50
user3939382 wrote: post #27

Earlier quoted context omitted.

I asked ChatGPT how to add a JSDoc type to a Vue 2 prop and it gave me a wrong answer. There have been several times where I’ve asked it questions and it sprinkles in well disguised misinformation. These tools are impressive but they definitely have limitations.

You're asking a model which is just a big unfocused and highly alignment taxed mush of the entire internet across all language, regardless of if its natural or computer, then (maybe) sliced across 8 experts. It can also simulate a Zizek vs Wittgenstein argument over Russian literature. the fact that it can usually write executing computer code is nothing short of magic to me. What one would want is something like LLa…

I think this concept will be the future of coding. You'll have something like CodeLLaMA-javascript trained specifically for the language, then you'll have LoRA-like files distributed with/for each library. I can image something like the definitely-typed project for typescript types, but instead of doing `yarn add @types/someLibrary` you'll be able to do `yarn add @loramodels/someLibrary`.

I'm javascript centric these days, but the same concept should work for most languages with a package manager.

Some of these features could even be integrated into say yarn/npm directly, you currently have a devDependencies section of a package.json..I could imaging having something along the lines of a "llmDependencies" section to define which main model and version to use and which "library" models to use.

Re: AI isn’t good enough

#183

Post by martindbp »

usrbinbash wrote: post #67
ambrozk wrote: post #18

This entire piece is based on one massive, unsupported assertion, which is that LLM progress will cease. Or, as the author puts it, "we are at the tail end of the first wave of large language model-based AI... [it] ends somewhere in the next year or two with the kinds of limits people are running up against." I want to know only one thing, which is what gives him the confidence necessary to say that. If that one stat…

> This entire piece is based on one massive, unsupported assertion, which is that LLM progress will cease. Which is countered by...the assertion that it won't? LLMs won't get intelligent. That's a fact based on their MO. They are sequence completion engines. They can be fine tuned to specific tasks, but at their core, they remain stochastic parrots. > I want to know only one thing, which is what gives him the confide…

If something stays in motion and has been so for some time, it's more important to explain why it would not continue rather than the default assumption that it will stop instantaneously. Show me a curve of diminishing returns and I'll believe you. If an object is in motion you'd need to show me that there is deceleration, or there is a wall just up ahead.

But the fact is that the loss goes down predictably with increased compute budget, data and model size (see Chinchilla Scaling Law). We've also seen that decreased loss suddenly results in new capabilities in discontinuous jumps. There is all reason to believe there is still some juice left in this scaling, exactly how far it can be taken is difficult to tell.

Re: AI isn’t good enough

#184

Post by lewhoo »

IanCal wrote: post #153
usrbinbash wrote: post #67

Earlier quoted context omitted.

> This entire piece is based on one massive, unsupported assertion, which is that LLM progress will cease. Which is countered by...the assertion that it won't? LLMs won't get intelligent. That's a fact based on their MO. They are sequence completion engines. They can be fine tuned to specific tasks, but at their core, they remain stochastic parrots. > I want to know only one thing, which is what gives him the confide…

> LLMs won't get intelligent. That's a fact based on their MO. They are sequence completion engines. A system that could perfectly predict what I would do in response to any particular stimuli, as a continuing sequence, would be exactly as intelligent as me. > They can be fine tuned to specific tasks, but at their core, they remain stochastic parrot Othello GPT was an attempt at answering this exact question, it's a…

A system that could perfectly predict what I would do in response to any particular stimuli, as a continuing sequence, would be exactly as intelligent as me.

That's certainly interesting but it's not a depiction of a LLM is it ? LLM's are not deterministic, and (perhaps) so are we so two non-deterministic systems can only occasionally align (or so I assume). Intuition says they may get "close enough", whatever that might be, and close enough is good enough in this case but I think you are making a giant assumption to the likes of since we can speed up matter to 1000km/h then IF we sped it up to light speed then ...[something]...

Re: AI isn’t good enough

#185

Post by ChatGTP »

jillesvangurp wrote: post #112

By the time AIs stop hallucinating, it will be effectively super human (i.e. much better) because humans hallucinate all the time. And some people are barely coherent. We're holding AIs to a much higher standard than ourselves. And we move the goal posts all the time as well. Let's deconstruct that title. "AI isn't good enough". Good enough for what? Great example of moving the goal posts. Because anytime it fails to…

"Guy calls humans flawed why claiming the same flawed species will build superhuman intelligence...", pretty good for a bunch of dumb LLMs in "meat bags"...

Re: AI isn’t good enough

#186

Post by pdimitar »

lynx23 wrote: post #179
pdimitar wrote: post #161

Earlier quoted context omitted.

You are ignoring the fact that a toddler, once their musculature develops, is able to learn to walk after several tries. Show me a humanoid robot that can do that.

A toddler capable of walking is still miles away from the topic at hand, a LLM capable of processing relatively complex text. It is well know that humans have a particularily long training period compared to other animals. So I dont understand why you are bringing up the walking-training. It seems quite unrelated.

Because humans learn from much less samples sizes than "AI". The overly optimistic point of view on "AI" is over-represented in HN and that makes the discussion unbalanced and puts nearly everyone in it in a filter bubble.

That's not OK.

Re: AI isn’t good enough

#187

Post by pdimitar »

Drakim wrote: post #166
pdimitar wrote: post #161

Earlier quoted context omitted.

You are ignoring the fact that a toddler, once their musculature develops, is able to learn to walk after several tries. Show me a humanoid robot that can do that.

To be fair a toddler's brain isn't "offline" while it's musculature develops, they kick and move their limbs constantly, and then later crawl and and even stand, before they finally take the first walking steps. It's just objectively wrong to say a toddler goes from zero to walking in "several tries".

Sure, not several. Not several trillions either though.

As mentioned in a sibling comment, HN anthropomorphizes AI too much. And is too optimistic about it. I just don't see the results and the value, people trip up all over themselves to congratulate themselves and the researches, yet 99.99% of the problems in the world persist.

I am one of these a-holes that wants to see results when money are invested. It still comes as a shock to some apparently.

Re: AI isn’t good enough

#188

Post by sebzim4500 »

pdimitar wrote: post #161
garblegarble wrote: post #158

Earlier quoted context omitted.

>Live systems in nature seem to solve similar problems with way less compute available Do they really? They're certainly more energy-efficient in business-as-usual mode, but a human brain has 86 billion neurons, 600+ trillion synapses(!), and each instance takes 15-20+ years to train to do complex logical tasks. Even if the per-cell work is tiny (and, is it? cells are amazingly complex), 86 billion (or 600+ trillion)…

You are ignoring the fact that a toddler, once their musculature develops, is able to learn to walk after several tries. Show me a humanoid robot that can do that.

Not sure about humanoids but there are RL algorithms that can learn to walk from scratch in a quadruped in minutes.

Re: AI isn’t good enough

#189

Post by datpiff »

squeaky-clean wrote: post #87
usrbinbash wrote: post #67

Earlier quoted context omitted.

> This entire piece is based on one massive, unsupported assertion, which is that LLM progress will cease. Which is countered by...the assertion that it won't? LLMs won't get intelligent. That's a fact based on their MO. They are sequence completion engines. They can be fine tuned to specific tasks, but at their core, they remain stochastic parrots. > I want to know only one thing, which is what gives him the confide…

> LLMs won't get intelligent. I think this sentence doesn't mean much unless we have a strict definition of what intelligence means. Just today ChatGPT helped me solve a DNS issue that I would not have been able to solve on my own in one day, let alone an hour. I'd consider it already more intelligent than myself when it comes to DNS.

> Just today ChatGPT helped me solve a DNS issue that I would not have been able to solve on my own in one day, let alone an hour. I'd consider it already more intelligent than myself when it comes to DNS.

Would you consider a search engine, or a book, to be as intelligent?

Re: AI isn’t good enough

#190

Post by usrbinbash »

glenstein wrote: post #115
usrbinbash wrote: post #67

Earlier quoted context omitted.

> This entire piece is based on one massive, unsupported assertion, which is that LLM progress will cease. Which is countered by...the assertion that it won't? LLMs won't get intelligent. That's a fact based on their MO. They are sequence completion engines. They can be fine tuned to specific tasks, but at their core, they remain stochastic parrots. > I want to know only one thing, which is what gives him the confide…

>Which is countered by...the assertion that it won't? No it's countered by principled restraint in not making an affirmative claim one way or the other. I've heard this referred to as the overconfident pessimism problem. Which is that normal, well founded scientific discipline and evidence-based restraint go out of the window when people declare, without evidence that they know certain advances won't happen. Because…

> when people declare, without evidence

The evidence for this, albeit empirical, is the history of AI development itself.

AI doesn't show continuous development over a long period of time. It always developed in steps. A new architecture or method is discovered and able to solve some previously hard or unsolveable problems.

Then this solution slowly develops in capability, mostly based on better and cheaper hardware, while it's quality plateaus.

It may be that LLMs will not follow that pattern. I don't think so, for reasons outlined above. But until it can be shown that they don't, that this really is the long-sought-after AI architecture that just gets better and better over time, I don't think that a healthy dose of pessimism is unwarranted based on history.

Post reply on HN