scruiser

Bonus: a recent comment is skeptical:

well, how do I play democracy with AI? It’s already 2025

We're already behind schedule, we're supposed to have AI agents in two months (actually we were supposed to have them in 2022, but ignore the failed bits of earlier prophecy in favor of the parts you can see success for)!

He made some predictions about AI back in 2021 that if you squint hard enough and totally believe the current hype about how useful LLMs are you could claim are relatively accurate.

His predictions here: https://www.lesswrong.com/posts/6Xgy6CAf2jqHhynHL/what-2026-looks-like

And someone scoring them very very generously: https://www.lesswrong.com/posts/u9Kr97di29CkMvjaj/evaluating-what-2026-looks-like-so-far

My own scoring:

The first prompt programming libraries start to develop, along with the first bureaucracies.

I don't think any sane programmer or scientist would credit the current "prompt engineering" "skill set" with comparison to programming libraries, and AI agents still aren't what he was predicting for 2022.

Thanks to the multimodal pre-training and the fine-tuning, the models of 2022 make GPT-3 look like GPT-1.

There was a jump from GPT-2 to GPT-3, but the subsequent releases in 2022-2025 were not as qualitatively big.

Revenue is high enough to recoup training costs within a year or so.

Hahahaha, no... they are still losing money per customer, much less recouping training costs.

Instead, the AIs just make dumb mistakes, and occasionally “pursue unaligned goals” but in an obvious and straightforward way that quickly and easily gets corrected once people notice

The safety researchers have made this one "true" by teeing up prompts specifically to get the AI to do stuff that sounds scary to people to that don't read their actual methods, so I can see how the doomers are claiming success for this prediction in 2024.

The alignment community now starts another research agenda, to interrogate AIs about AI-safety-related topics.

They also try to contrive scenarios

Emphasis on the word"contrive"

The age of the AI assistant has finally dawned.

So this prediction is for 2026, but earlier predictions claimed we would have lots of actually useful if narrow use-case apps by 2022-2024, so we are already off target for this prediction.

I can see how they are trying to anoint his as a prophet, but I don't think anyone not already drinking the kool aid will buy it.

I think Eliezer has still avoided hard dates? In the Ted talk, I distinctly recall he used the term "0-2 paradigm shifts" so he can claim prediction success for stuff LLMs do, and paradigm shift is vague enough he could still claim success if its been another decade or two and there has only been one more big paradigm shift in AI (that still fails to make it AGI).

Is this the corresponding lesswrong post: https://www.lesswrong.com/posts/TpSFoqoG2M5MAAesg/ai-2027-what-superintelligence-looks-like-1 ?

Committing to a hard timeline at least means making fun of them and explaining how stupid they are to laymen will be a lot easier in two years. I doubt the complete failure of this timeline will actually shake the true believers though. And the more experienced ~~grifters~~ forecasters know to keep things vaguer so they will be able to retroactively reinterpret their predictions as correct.

I can already imagine the lesswronger response: Something something bad comparison between neural nets and biological neurons, something something bad comparison with how the brain processes pain that fails at neuroscience, something something more rhetorical patter, in conclusion: but achkshually what if the neural network does feel pain.

They know just enough neuroscience to use it for bad comparisons and hyping up their ML approaches but not enough to actually draw any legitimate conclusions.

Galaxy brain insane take (free to any lesswrong lurkers): They should develop the usage of IACUCs for LLM prompting and experimentation. This is proof lesswrong needs more biologists! Lesswrong regularly repurpose comp sci and hacker lingo and methods in inane ways (I swear if I see the term red-teaming one more time), biological science has plenty of terminology to steal and repurpose they haven't touched yet.

Yeah there might be something like that going on causing the "screaming". Lesswrong, in it's better moments (in between chatbot anthropomorphizing), does occasionally figure out the mechanics of cool LLM glitches (before it goes back to wacky doom speculation inspired by those glitches), but there isn't any effort to do that here.

I agree. There is intent going into the prompt fondler's efforts to prompt the genAI, it's just not very well developed intent and it is using the laziest shallowest method possible to express itself.

If you understood why the splattered paint was art, you would also understand why the AI generated images aren't art (or are, at best, the art of hacks). It seems like you understand neither.

Another episode in the continued saga of lesswrongers anthropomorphizing LLMs to an absurd extent: https://www.lesswrong.com/posts/MnYnCFgT3hF6LJPwn/why-white-box-redteaming-makes-me-feel-weird-1

Lol, Altman's AI generated purple prose slop was so bad even Eliezer called it out (as opposed to make a doomer-hype point):

Perhaps you have found some merit in that obvious slop, but I didn't; there was entropy, cliche, and meaninglessness poured all over everything like shit over ice cream, and if there were cherries underneath I couldn't taste it for the slop.