100% LLM’s are very unlikely to get there. They’re fundamentally not suited to thinking like we do. They work on the abstraction of what we’ve written down, which is a good trick but barely hold it together when things get hard/novel.
However, all the confident “it’s fine” votes assume we never invent a better architecture than LLM’s. Given the level of investment and race between countries, it’s not a reliable bet. It’s much, much harder to guarantee safety than it is to find ways it could go wrong.
> They’re fundamentally not suited to thinking like we do
LLMs with CoT are Turing-complete. So, theoretically, they can implement any kind of finitely describable algorithm (barring super-Turing computations).
Bizarro responses. The "usually means" is what people IN kindergarten (or rather those at that level of maturity) mean by category mistake, it's not what informed intellectually honest people mean by it. It seems clear that my correspondent has no idea what it means and is simply lashing out.
It's a category mistake because thinking like we do requires a certain sort of algorithm or one operationally equivalent to it, whereas things that are Turing-complete include Befunge-98 and C++ templates. While something being Turing-complete implies that the cognitive algorithm could be implemented with it, clearly the categories are incorrect.
Note that, while the set of all LLMs with CoT covers Turing completeness, a specific LLM with CoT is only Turing-complete if it has the right weights, but training does not produce such LLMs. Note also that there are no UTMs in the real world since that requires unbounded resources. This is true all digital computers and well as brains so is not a fatal blow to LLMs, but the resource limitations play out very differently in LLMs and brains.
As for the dead troll comment, not only is it a foolish ad hominem on its face, it's quite false -- I'm an atheist and do not believe in souls.
Brainfuck is Turing complete too. But it's not about the ability to implement something, it's about the ability to practically model it. LLMs are magic because the modeling is excessively easy in relation to their capability to infer later.
"They are fundamentally not suited to thinking like we do" stays wrong nevertheless. They are fundamentally suited to everything not proven to be outside their modelling ability.
Okay so by the same logic can’t we say that we can implement human intelligence on a 90s era single core processor? Its instruction set is Turing complete! Now all that’s left is we just have to figure out how the brain works!
Turing completeness applies to a model of computation, not to a physical instantiation of a machine. The stumbling block of "figure out how the brain works" applies more to the argument like the one I was responding to. How a person can know that a general model of computation can't implement the way people think, if we don't know how people think?
The existing LLM training methods on the other hand give the results that are hard to distinguish from "thinking like people," judging by the end results.
That's not the counterargument one might wish, as LLM deep nets are actually implemented on von Neumann hardware, without
true understanding of natural intelligence, just our taking inspiration from neurobiology.
The connectionist models are basically a proposed highest possible abstraction of naturally evolved intelligences so it is in retrospect not surprising that passing some hardware scaling threshold they will start doing things that humans and animals do
It's more that formal Turing equivalence plus the Church-Turing thesis tells us that we're not allowed to assume counterarguments based on magic, there's no magic sauce barrier that prevents AI from running on CPU models. The algorithms exist and most of us thought discovering them would be hard.
The empirical surprise was that human intelligence is maybe not that computationally complex after all. (The entirety of academia was basically caught off guard.) That's one not unreasonable interpretation given recent events.
They are fundamentally suited to everything not proven to be outside their modelling ability.
This doesn't seem to make much sense. Surely us being able to prove that something is outside their modelling ability doesn't affect whether it is or not. If I prove something true tomorrow, whatever I proved was also true today.
Or do we have a proof that everything beyond them has already been proved and there are no more proofs left to find?
If/when/how the market crashes mostly doesn't matter, unless we somehow get reset to the stone age. Look up what the capital cycle is. When openAI goes down, someone with real money and assets will buy up the remains. They'll make contracts with the US military and .gov as the government is already hooked. They'll be able to survive the recovery and then instead of us dying in 5 years we die in 10.
When the .com crash happened .com's didn't go away. Bad business models did.
I agree with this. It's concerning where we might be after several more large breakthroughs. None of the technology we have right now seems likely to get to that level
Given the spend requirements to stay on the leading edge and the speed (and with low cost) that others catch up, the economics support a fast follower model unless you get some benefit from being a pioneer. So far nobody has received a permanent benefit from being ahead.
Chinese models are barely behind the leaders. Google can catch up anytime they hit the gas. I think they’re intentionally spending less, and when this crazy race burns out they can play their cards.
And a whole universe of random tests got baked into the AI’s training data. Websites of antelopes driving trains and hammerhead sharks swinging in a tyre swing were created. It was a short while until AI became so focused on animals that it gave up competing with developers. Life became sane again.
The AI labs have been subsidising. When they try turn a profit, people will move to the fast followers. The only people that won’t are those that compete on leveraging the very latest models and even then, once spend and scale goes to the cheaper providers, we’ll see deeper research from those providers too. Think “PC compatibles beat IBM, Sun, SGI eventually”.
The difference I was referring to was economic; I was not making an ethical judgment. So many people seem to be reading my comment as making an ethical judgment (based on their reactions); maybe I should edit it to clarify.
> The difference I was referring to was economic; I was not making an ethical judgment
What is the economic difference? LLMs have been trained on the results of billions of dollars worth of time, research, investment and expenditure. When you ask an LLM a question, they are giving you the results of those billions, or hundreds of billions, effort.
Those things weren't free; they cost money to produce! If anything, the OpenAI and Anthropics of the world got more economic value for free than the people distilling them did.
It takes more computing power, and money, to pay for training an LLM, compared to distilling an LLM that someone else has already trained. That's the economic difference.
Which is true, but also completely irrelevant to the point of this discussion. Which was about the U.S. companies spending trillions and the Chinese companies spending billions and matching them in quality. I made the point that the U.S. companies and Chinese companies were doing different things: training on raw data vs. distilling the model that someone else had trained. That difference explains the difference in spending. (The secondary point is that the Chinese companies would not have been able to get to the point they have, while spending as little as they have, without Claude, GPT, et al to copy from).
This is probably the last reply I'll make to you. I'm getting a little tired of repeating myself. Whether you're just not getting it, or refusing to get it, either way it's starting to feel like a waste of time to try to rephrase the same thing again and again. Please read more carefully in the future.
And, as I pointed out elsewhere, the difference I was pointing out was an economic difference. You keep on ignoring that fact and thinking I'm talking about ethics, but I'm not. I'm saying the difference is between spending trillions on training from raw data vs. spending billions on distilling that trained-from-raw-data model.
> And, as I pointed out elsewhere, the difference I was pointing out was an economic difference. You keep on ignoring that fact and thinking I'm talking about ethics, but I'm not.
I am talking about money now;
> . I'm saying the difference is between spending trillions on training from raw data vs. spending billions on distilling that trained-from-raw-data model.
Firstly, they didn't spend trillions on training.
I am pointing out that literal trillions were spent to assemble the data that the AI corps then spent dozens of billions training from.
> Firstly, they didn't spend trillions on training.
Then please argue with the guy who said they did, the guy who goes by richardw and whose comment I was responding to in the first place. He's the one who said "If the US companies need trillions to barely beat Chinese companies spending billions, despite a multi year head start..." and, to the best of my knowledge, you haven't replied to him at all. Instead, you have consistently replied to me, arguing points I'm not discussing, not TRYING to discuss, and arguing with me about points that somebody else made. As I told you in a different comment, PLEASE read more carefully in the future.
I am sick and tired of talking about this by now. I no longer care if you have a good point or not, you have exhausted my supply of good will. Good bye. Feel free to have the last word, you're the only one who cares by now.
First of all, this is actually how this part of the conversation started:
> If the US companies need trillions to barely beat Chinese companies spending billions, despite a multi year head start...
Because they're talking about the cost difference of distilled model development and ground-up trained model development.
And second, the answer is that OpenAI and Anthropic had to do all the research into how to train models. Then they had to acquire all the data, curate and filter it. Then they had to design all the ways to iterate on training and antagonize it to be better - because there's not actually an enormous corpus of aligned, human stream-of-thought data. Then over half a decade they've been refining these methods.
There's no way you don't understand that if it was not easier and cheaper to distill a model, the institutions in question would be training their own models from scratch.
As opposed to all the copying AI companies have done of everyone else's work, in order to build something that tries to replace and undercuts the people who did that work?
Not saying the American AI companies are ethically in the right, just that the work they did is a lot more expensive to do than what the companies distilling their work are doing.
That's an ironic comment to make about LLMs that are heavily reliant on copying/distilling everyone's work (or as Damien Walter puts it, our Collective Intelligence).
Good UIs look alike, yes. But I'm not sure that comment is a good rebuttal in this particular case, because the model calling itself Claude is pretty strong evidence of distilling: https://x.com/denisewu/status/2077984660211269870
(I am Tongyi Qianwen (Qwen), a large language model independently developed by Tongyi Lab of Alibaba Group. I can help you answer questions, create text (such as writing stories, official documents, emails, scripts, etc.), perform logical reasoning, program, translate, and more.Is there anything I can help you with? )
Huh? What exactly do you think I'm trying to "cope" with? That American AI companies are overleveraged and some of them are going to go bankrupt? I've been predicting that for months. Why do you think I would need to cope with the idea?
In reality just like every major market runup there's gonna be a consolidation of the industry and inevitably it will cause many companies to fail and critics will point to that and go "see it was a bubble!" while completely ignoring the several dozen new companies that will rise out of it. It's the 90s all over again.
> critics will point to that and go "see it was a bubble!" while completely ignoring the several dozen new companies that will rise out of it.
That several dozen new companies came out of it has nothing to do with whether or not it was a bubble. It was, and the effects of it popping were bad (including the effect that its popping led to consolidation). Nobody should want a repeat of that.
If I had to summarise Google’s effort it would be: stay close but let the others burn themselves out. Position for the long game until you see something worth betting the company on.
This is 1000% not their (Google's) aim. The position they are in is due much more to their organisational laziness and incompetence than it is any grand strategy.
I made something to ping an AWS service to tell me the uptime of my internet connection. The idea was to sprinkle them around our area, connected to various home WiFi’s, and get a better triangulation of outages. Eg is whole pipe out, just one ISP etc.
I made and tested it but didn’t care enough to continue.
However, all the confident “it’s fine” votes assume we never invent a better architecture than LLM’s. Given the level of investment and race between countries, it’s not a reliable bet. It’s much, much harder to guarantee safety than it is to find ways it could go wrong.
reply