what?

do you think we live in the alien universe or something? I seriously sometimes try to argue with some of these comments but jesus christ

If you give me ANY idea of why you think my argument is stupid, we could actually discuss it?

My company is building an agentic layer in its product, google and the other hyperscalers do that too.

I already want to get rid of one person in my team because he was slow before but was able to contribute something, now I want to have his position as AI tokens.

you think 10% of the entire usa market is gonna go to anthropic and I'm supposed to somehow find a way with words for you to understand how stupid that sounds?

some of you are acting like earth is the size of jupiter and there's 800 billion humans not 8

I've been trying to generate some product videos with opus 5.5 and I use the API because I don't like subscriptions.

I burned about $150 of tokens over a few hours of chatting, although opus 5.5 did generate some 10-30kb html files at a time, containing the "videos".

I think subscriptions are extremely subsidized still and what happened yesterday was just a small dip of what's to come.

Dear god, you can use sonnet for that.

opus was the only one who did well, argon too

you don't even know my specifications, why are you parroting something useless?

LLMs can't do a backflip. I can.

Unitree's H2-X can.

https://youtu.be/Y3zb4o47yqA

is unitree h2-x an llm?

why not? I'm sure you could bolt one onto it...

I can’t do a backflip. I guess I’m subservient to LLMs then.

the equivalency is both go bankrupt if people win as much as the house wants you to think they win

Prediction markets' business model is taking transaction fees from contracts between independent counterparties. Kalshi won't go bankrupt no matter how much you make there (though 'people' collectively will make negative [fees + taxes]).

Kalshi does make both sides feel like they'll probably win, and if that actually happened, Kalshi would definitely lose money.

I don't know why I seem to be the only one to wonder this but .. even if AI hypothetically can output smarter data to train a better AI, how do we know for sure this loop just goes on without eventually plateauing? What if what happens is just .. the data becomes more high definition, and then it's just the same iteration of the same "high" definition.

We don't know.

That is, however, the argument at its core. We don't know where RSI would lead. It could produce something roughly human-intelligent. It could produce something slightly smarter than a human. It could produce something radically, radically smarter than a human. We don't know. A lot of people suspect the correct scale would be a logarithmic one, which seems to conform to the pace of improvement being seen, but that's not something we know either.

It could also produce any of the above, but with a reasonably human architecture to the resulting intelligence, or an architecture radically divergent from ours. Although which of those is "better" or "worse" is a rather open question.

Something about OpenAI chasing math problems is telling me they're after a better architecture because they've seen model collapse over and over.

This is indeed part of the grand illusion in the field - that improvements in cognitive capabilities translate linearly into research progress. There are talks where experts report projections of RSI progress growing exponentially.

This is why AI companies use research in a deterministic and rigorous field (mathematics) as a benchmark - it creates the impression that humanity will achieve comparable progress in other fields.

Having said like, like for other fields (e.g. graphics), hardware capacity slowly increases, and new algorithms are devised, so there will hardly be a strict plateau.

I'm not implying that research progress will be necessarily very slow, but that it's severely misrepresented for sure.

It even seems to me that plateau-ing is an inevitability. Unless there is some way to reach AGI which you could deduce from a combination of ideas which already exist in the training dataset, but that seems rather unlikely.

The reality of "RSI" is just attempting to automate the jobs of people working on building/training LLMs - it's not about creating "smarter data". Much post-training nowadays doesn't even use data - it uses thousands of custom training environments for agents to attempt to do better in.

Digital eugenics, phrenology, etc. Humans constantly think they can identify a optimum fit function for humans in society. Because thats easier making a society optimum for everyone.

This is no different. Its like a single ant trying to redefine what a ant is. Typically with catastrophic results.

I don't think the idea is better data so much as algorithmic improvements.

The universe presents an endless amount of training data. Some if it, like maths, is abstract; you can synthesize the data.

Yes, I'm quite confident that's what OpenAI is doing. These math proofs are just a side effect of them trying to create new data out of the blue for a better model. But that's still chasing a wild goose and hitting jackpots with 10,000 active agents, it doesn't necessarily prove RSI.

RSI is simply what you get when human involvement tends to zero. Once they set up the training loop you get RSI. I am strongly opposed to this.

makes me wonder if these retail investors are noticing how everyone keeps mentioning them about holding the bag

I think OpenAI's limitations are quickly verifiable results. 10,000 agents running in parallel aren't exactly talking to each other. They have a goal and one eventually hits the goal which can be automatically confirmed. This is why math is currently attacked, lean is easy to confirm A = B.

They aren't gonna create room temperature superconductors because they'll need a lab that can quickly, and verifiably, test superconductors at 10,000 per minute and that's just a physical limitation no matter how many datacenters you throw at it.

Art is the ability to transform one experience into another. That's a generalization only RSI can hopefully come up with.