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

Evolving OpenAI's Structure

openai.com

Post reply on HN
Print view View original

361–370 of 746 posts

Re: Evolving OpenAI's Structure

#361

Post by devjab »

orionsbelt wrote: post #261
caseyy wrote: post #222

Earlier quoted context omitted.

It's doubtful if there even is a race anymore. The last significant AI advancement in the consumer LLM space was fluent human language synthesis around 2020, with its following assistant/chat interface. Since then, everything has been incremental — larger models, new ways to prompt them, cheaper ways to run them, more human feedback, and gaming evaluations. The wisest move in the chatbot business might be to wait and…

Saying LLMs have only incrementally improved is like saying my 13 year old has only incrementally approved over the last 5 years. Sure, it's been a set of continuous improvements, but that has taken it from a toy to genuinely insanely useful. Personally, deep research and o3 have been transformative, taking LLMs from something I have never used to something that I am using daily. Even if the progress ends up plateaui…

> any company sitting this out risks being unable to capture users back from Open AI at a later date.

Why? I paid for Claude for a while, but with Deepseek, Gemini and the free hits on Mistral, ChatGPT, Claude and Perplexity I'm not sure why I would now. This is anecdotal of course, but I'm very rarely unique in my behaviour. I think the best the subscription companies can hope for is that their subscribers don't realize that Deepseek and Gemini can basically do all you need for free.

Re: Evolving OpenAI's Structure

#362

Post by latentsea »

dom96 wrote: post #242
voidspark wrote: post #219

Earlier quoted context omitted.

> AGI would mean something which doesn't need direction or guidance to do anything There can be levels of AGI. Google DeepMind have proposed a framework that would classify ChatGPT as "Emerging AGI". ChatGPT can solve problems that it was not explicitly trained to solve, across a vast number of problem domains. https://arxiv.org/pdf/2311.02462 The paper is summarized here https://venturebeat.com/ai/here-is-how-far-we…

This constant redefinition of what AGI means is really tiring. Until an AI has agency, it is nothing but a fancy search engine/auto completer.

I agree. AGI is meaningless as a term if it doesn't mean completely autonomous agentic intelligence capable of operating on long-term planning horizons.

Edit: because if "AGI" doesn't mean that... then what means that and only that!?

Re: Evolving OpenAI's Structure

#363

Post by lukan »

always_imposter wrote: post #128
voidspark wrote: post #96

Earlier quoted context omitted.

Their multimodal models are a rudimentary form of AGI. EDIT: There can be levels of AGI. Google DeepMind have proposed a framework that would classify ChatGPT as "Emerging AGI". https://arxiv.org/abs/2311.02462

AGI would mean something which doesn't need direction or guidance to do anything. Like us humans, we don't wait for somebody to give us a task and go do it as if that is our sole existence. We live with our thoughts, blank out, watch TV, read books etc. What we currently have and possibly in the next century as well will be nothing close to an actual AGI. I don't know if it is optimism or delusions of grandeur that d…

It seems like you believe AGI won't come for a long time, because you don't want that to happen.

The turing test was succesfull. Pre chatGPT, I would not have believed, that will happen so soon.

LLMs ain't AGI, sure. But they might be an essential part and the missing parts maybe already found, just not put together.

And work there will be always plenty. Distributing ressources might require new ways, though.

Re: Evolving OpenAI's Structure

#364

Post by horhay »

atleastoptimal wrote: post #184
grey-area wrote: post #142

Earlier quoted context omitted.

Well I think you’re correct that they know the jig is up, but I would say they know the AI bubble is about to burst so they want to cash out before that happens. There is little to no money to be made in GAI, it will never turn into AGI, and people like Altman know this, so now they’re looking for a greater fool before it is too late.

AI companies are already automating huge swaths of document analysis, customer service. Doctors are straight up using ChatGPT to diagnose patients. I know it’s fun to imagine AI is some big scam like crypto, but you’d have to be ignoring a lot of genuine non hype economic movement at this point to assume GAI isn’t making any money. Why does the forum of an incubator that now has a portfolio that is like 80% AI so rou…

Lol they are not using ChatGPT for the full diagnosis. They're used in steps of double checking knowledge like drug interactions and such. If you're gonna speak on something like this in a vague manner I'd suggest you google this stuff first. I can tell you for certain that that part in particular is a highly inaccurate statement.

Re: Evolving OpenAI's Structure

#365

Post by m3kw9 »

This sounds like a good middle ground between going full capitalism and non-profit. This way they can still raise money and also have the same mission, but a weakened one. You can't have everything.

Re: Evolving OpenAI's Structure

#366

Post by voidspark »

JumpCrisscross wrote: post #355
voidspark wrote: post #337

Earlier quoted context omitted.

> Gluten-free means free of gluten. Bad analogy. That's a binary classification. AGI systems can have degrees of performance and capability. > Humans are not. LLMs are. My point is that if you oversimplify LLMs to "word autocompletion" then you can make the same argument for humans. It's such an oversimplification of the transformer / deep learning architecture that it becomes meaningless.

> That's a binary classification. AGI systems can have degrees of performance and capability The "g" in AGI requires the AI be able to perform "the full spectrum of cognitively demanding tasks with proficiency comparable to, or surpassing, that of humans" [1]. Full and not full are binary. > if you oversimplify LLMs to "word autocompletion" then you can make the same argument for humans No, you can't, unless you're p…

Then you are simply rejecting any attempts to refine the definition of AGI. I already linked to the Google DeepMind paper. The definition is being debated in the AI research community. I already explained that definition is too limited because it doesn't capture all of the intermediate stages. That definition may be the end goal, but obviously there will be stages in between.

> No, you can't, unless you're pre-supposing that LLMs work like human minds.

You are missing the point. If you reduce LLMs to "word autocompletion" then you completely ignore the the attention mechanism and conceptual internal representations. These systems have deep learning models with hundreds of layers and trillions of weights. If you completely ignore all of that, then by the same reasoning (completely ignoring the complexity of the human brain) we can just say that people are auto-completing words when they speak.

Re: Evolving OpenAI's Structure

#367

Post by brendoelfrendo »

dr_dshiv wrote: post #354
Night_Thastus wrote: post #74

Earlier quoted context omitted.

Nothing OpenAI is doing, or ever has done, has been close to AGI.

https://www.noemamag.com/artificial-general-intelligence-is-... Here is a mainstream opinion about why AGI is already here. Written by one of the authors the most widely read AI textbook: Artificial Intelligence: A Modern Approach https://en.wikipedia.org/wiki/Artificial_Intelligence:_A_Mod...

I would argue that this is a fringe opinion that has been adopted by a mainstream scholar, not a mainstream opinion. That or, based on my reading of the article, this person is using a definition of AGI that is very different than the one that most people use when they say AGI.

Re: Evolving OpenAI's Structure

#368

Post by Yizahi »

jonfromsf wrote: post #322
foobiekr wrote: post #81

Earlier quoted context omitted.

I think this is right but also missing a useful perspective. Most HN people are probably too young to remember that the nanotech post-scarcity singularity was right around the corner - just some research and engineering way - which was the widespread opinion in 1986 (yes, 1986). It was _just as dramatic_ as today's AGI. That took 4-5 years to fall apart, and maybe a bit longer for the broader "nanotech is going to ch…

Every consumer has very useful AI at their fingertips right now. It's eating the software engineering world rapidly. This is nothing like nanotech in the 80s.

Sure. But fancy autocomplete for a very limited industry (IT) plus graphics generation and a few more similar items, are indeed useful. Just like "nanotech" coating of say optics or in the precise machinery or all other fancy nano films in many industries. Modern transistors are close to nano scale now, etc.

The problem is that the distance between a nano thin film or an interesting but ultimately rigid nano scale transistor and a programmable nano level sized robot is enormous, despite similar sizes. Same like the distance between an autocomplete heavily relying on the preexisting external validators (compilers, linters, static code analyzers etc.) and a real AI capable of thinking is equally enormous.

Re: Evolving OpenAI's Structure

#369

Post by tbrownaw »

JumpCrisscross wrote: post #293
lanthissa wrote: post #65

Earlier quoted context omitted.

AGI can't really be a winner take all market. The 'reward' for general intelligence is infinite as a monopoly and it accelerates productivity. Not only is there infinite incentive to compete, but theres decreasing costs to. The only world in which AGI is winner take all is a world in which it is extremely controlled to the point at which the public cant query it.

> AGI can't really be a winner take all market. The 'reward' for general intelligence is infinite as a monopoly and it accelerates productivity The first-mover advantages of an AGI that can improve itself are theoretically unsurmountable. But OpenAI doesn't have a path to AGI any more than anyone else. (It's increasingly clear LLMs alone don't make the cut.) And the market for LLMs, non-general AI, is very much not w…

> The first-mover advantages of an AGI that can improve itself are theoretically unsurmountable.

This has some baked assumptions about cycle time and improvement per cycle and whether there's a ceiling.

Re: Evolving OpenAI's Structure

#370

Post by henryfjordan »

dr_dshiv wrote: post #354
Night_Thastus wrote: post #74

Earlier quoted context omitted.

Nothing OpenAI is doing, or ever has done, has been close to AGI.

https://www.noemamag.com/artificial-general-intelligence-is-... Here is a mainstream opinion about why AGI is already here. Written by one of the authors the most widely read AI textbook: Artificial Intelligence: A Modern Approach https://en.wikipedia.org/wiki/Artificial_Intelligence:_A_Mod...

Why does the Author choose to ignore the "General" in AGI?

Can ChatGPT drive a car? No, we have specialized models for driving vs generating text vs image vs video etc etc. Maybe ChatGPT could pass a high school chemistry test but it certainly couldn't complete the lab exercises. What we've built is a really cool "Algorithm for indexing generalized data", so you can train that Driving model very similarly to how you train the Text model without needing to understand the underlying data that well.

The author asserts that because ChatGPT can generate text about so many topics that it's general, but it's really only doing 1 thing and that's not very general.

Post reply on HN