The exponentials won't save you either. In fact it's more likely that a sigmoid takes over at some point.
The sigmoids won't save you
221–230 of 297 posts
Re: The sigmoids won't save you
#222AI has scaled well according to convenient measures. It (neural networks) have the property that whatever you define, they can rapidly be trained master it. We’re able to show that various tasks of increasing complication do not require intelligence and can be framed as autoregressive RL problems. I personally don’t think AI is any closer to sentient intelligence than LeNet; it’s almost trivially clear, we know how i…
Is it not the case that a lot of the recent gains are just the coding harness that directs the LLM? That coding harness isn’t all that intelligent, simple pattern matching that maps to well defined tasks a programmer might do.
Re: The sigmoids won't save you
#223Forecasts are a thought exercise, not the revelation of something foretold. Best thing to do is think of the outcome you wish for and then try to take whatever actions you can to help make it so. Like with climate change for example.
Re: The sigmoids won't save you
#224FYI: The author has predicted that "AGI" will be here in 1-2 years and has staked his public reputation on it. He is personally invested in trendlines being lindy rather than sigmoid. I don't think you can use lindy on trends as if trends are static objects, but that's another conversation.
If I'm not mistaken, he's either affiliated with or otherwise connected to the effective altruist movement, hence he can't be unbiased. I find this article tells an interesting perspective on it: https://www.noemamag.com/the-politics-of-superintelligence/
Re: The sigmoids won't save you
#225Black Swan Events probably look like nothing just before they happen.
Re: The sigmoids won't save you
#226I felt the better takeaway from this was that it's impossible to know for certainty how long this will or will not continue regardless of the data or models you're using, because if you (or anyone else) could predict that accurately they'd be one of the richest people on the planet. I don't know when (or if) AI will implode or succeed with any degree of provable certainty, because that's not my area of expertise. Rat…
I think his agenda here is to point out that your probability distribution for AI outcomes should be broad (what you said), but most importantly: this means you must take seriously the possibility that we are gonna get superintelligence quite soon. Basically a lot of people say "but isn't it also pretty likely that we DON'T get superintelligence?" And, yes, it is. But superintelligence being even a remotely plausible…
You want to go to the store to get ice cream. Ice cream is delicious and the value of eating ice cream is a small positive, let's say x. There's a one in ten million chance you'll get hit by a car on the way and die, and your life is infinitely precious, therefore the expected value of going is x times 1 = x, and the one of not going is 1/10m times negative infinity which is negative infinity. You are a rational person, so you don't go. In fact you don't do much of anything. Your value model of every activity has collapsed to a single value.
That's the problem with 'singularity' arguments. The people making them ignore the fact that the mathematical definition of the word means 'the model of outcomes collapses to a single value' therefore the model stops being useful, yet they somehow claim to be able to make predictions beyond the singularity. It's like those shitty Facebook math posts where they divide both sides of the equation by 0 (the fact hidden by some sleight of hand), to 'prove' that 2=1.
The formulation of the singularity involves putting outrageous values into the parameters of the model of reality, and denominator ignorance, and then claiming 'rationally' determining that the consequences are too severe to ignore.
Re: The sigmoids won't save you
#227Earlier quoted context omitted.
Yet the evidence is on the side of the hype? We don’t see any mechanism or cogent framework for what limits exist here theoretically that I’m aware of, are you? Epoch had a great article a year ago looking at several bottlenecks in terms of scale and back then we were about 4 orders of magnitude away from hitting them. We’re probably now closer to 3. Yet scale is only part of the performance equation, a fairly big ch…
Why is the evidence on the side of the hype? Why do you assume something is size X just because nobody has proved it's smaller yet? The evidence is just whatever it is - we cannot make predictions with it.
We have very stable trends in observed performance in a LOT of different measurements and stable scaling laws that continue to hold true across at the largest scales we have built. These scaling laws have a reasonable theoretical justification and make predictions which continue to hold true. Pretraining perplexity does seem related to downstream measures of performance but those too show very stable trends in performance gains as well, even if you look at these in isolation. Look at the epoch capability index as a good summary statistic across a number of benchmarks.
So: yes we can and do make predictions with it and that’s how we get funding internally and externally to build at these scales.
Bottlenecks arise about 3 orders of magnitude from now.
On the other hand, any particular justification you have in mind for your point?
Re: The sigmoids won't save you
#228I think an interesting thing about recent AI developments is that its all happening right as we hit the diminishing returns side of another "exponential that's actually a sigmoid" which is Moore's law. The naive expectation is that AI will slow down b/c Moore's law is coming to an end, but if you really think about the models and how they are currently implemented in silicon, they are still inefficient as hell. At so…
Moore's law is bypassed with volume--more datacenters
The argument I've heard about how special Moore's law is that if you take a baby crawling at 10cm/s and 1000x it, you get almost Mach 3, which is almost as fast as the SR71 Blackbird flies. yet with a 18 month cadence, that's 15 years of progress, which has happened multiple times during the existence of the VLSI industry.
Re: The sigmoids won't save you
#229I felt the better takeaway from this was that it's impossible to know for certainty how long this will or will not continue regardless of the data or models you're using, because if you (or anyone else) could predict that accurately they'd be one of the richest people on the planet. I don't know when (or if) AI will implode or succeed with any degree of provable certainty, because that's not my area of expertise. Rat…
Which doomer argument have you found what problem with?
Re: The sigmoids won't save you
#230Earlier quoted context omitted.
So, this is not quite right: Alexander contributed to the report, but his personal opinion is more like the mid-2030s[1]. Freddie feels like this is him backing down from the original statement, but in fact he said this at the time the report was published, and in fact pointed out a graf below the quote that Freddie claims does tie him to 2027: > Do we really think things will move this fast? Sort of no - between the…
AI boosters really are detached from reality. LLMs are nothing close to AGI and not going to lead to it, they can’t distinguish right from wrong, they can’t count, they can’t reason, they generate plausible text from a vast databank of connected text. Apparently that is enough to fool many people but it’s nothing close to AGI which would require internal models of the world, reasoning etc. We are nowhere close to AGI…
> LLMs are nothing close to AGI and not going to lead to it, they can’t distinguish right from wrong, they can’t count, they can’t reason, they generate plausible text from a vast databank of connected text.
Argument?
Are LLMs close to being able to significantly help AGI researchers?