Hacker News Reader

Comment by state_less | Hacker News Reader

Even before LLMs, ML folks were already aware that you can use a model to teach another model. I doubt this is something AI companies put at the top of their investor materials, but it's been nice to see it play out.

That said, there are other moat factors like, a US company needing to use a US AI provider, sticky customers due to corporate onboarding friction, and others. Not nothing, but not as large a moat as some imagined.

View on HN

Yes, but 10 or 15 years ago, I would have thought that there'd be more to it than just a slight modification on the ideas behind a CNN to get this level of AI.

There were somewhat good reasons to think it needed more than just this data-driven ML approach.

There's something startling about how (relatively) simple these networks are and yet how powerful they are. The main ingredient the AI darlings are using is vast amounts of compute and data. I don't want to take away anything from what the researchers came up with, but I suspect even they are surprised at how capable some of these models have become.

early on there was a lot of talk about "emergent behaviors" in the models where they were good at things that were unexpected or did not align to the training data. IIRC doing arithmetic is one example from early on. I think this is where the AGI craze took off, the labs were throwing more and more data in the training to see what other behaviors would emerge. The thought was with enough data and enough parameters AGI would surface on its own.

Then i think tool use became a priority or at lest a sibling priority to more data/more params. Along with multiple specialized models communicating with each other which is sort of a special case of tool use. That pretty much brings us to today.

this comment is delusional. LLMs are awful at arithmetic and AGI is still very sci-fi, otherwise Claude would have told Anthropic how to cheaply generate energy for it to justify its existence by now. As long as the energy use debate persists you can be assured AGI has not arrived.

I feel you're conflating AGI and ASI here.

I'm unnerved by how alphazero is more complicated than the "intelligent llms"; it has at least multiple heads and MCTS, a search algorithm. The LLMs seem to just be monolithic (if complicated) architectures where tokens go in the bottom and tokens are spit out at the top.

imagine what would happen if you gave it MCTS and the data that LLMs were trained on

Reply on news.ycombinator.com