runeks

Except that we have a clear definition of what a computer is but not of what constitutes intelligence

Go on...

Their demo could also change goals, e.g. avoid monsters instead of shooting them, by changing some variable. I'm sure that could also be used to change goals.

> economic rent

What's the definition of this term?

> The tech group has issued a force majeure notice related to the power supply to the project developer, a unit of Blue Owl Capital, after the site ran into permitting delays after rising local opposition.

Since when can failing to get a permit be considered force majeure?

Doesn't majeure specifically denote something large scale and out of anyone's control, such as a hurricane?

Because we import the agricultural products we consume from other countries?

> With over 57 million sold worldwide since it went on sale in 1966, the Toyota Corolla is widely considered the best-selling vehicle to date.

Is it or is it not the best-selling vehicle to date? Should be pretty simple to determine.

> It regularly seems to come up with terms and descriptions for things in its chain of reasoning and then uses these terms in its output assuming you understand what it's talking about.

GPT does this all the time, too (both Sol and Astra). I constantly have to tell it to not use terms that were not part of the initial prompt.

> In the post-AGI economy, humans will eventually stop writing and reading code, but we still need an ambiguity-free way to tell the AIs building the world around us what we want done.

> With laws, our intents can be much more precise than natural language.

Doesn't this just mean that the code is now "laws", ie. the code is now the spec.

Given this, is there any reason think that writing the "laws" for a complex system is any easier than writing the old-fashioned code that implements it?

No, it doesn't assume that. It simply assumes that you can verify whether or not the LLM's implementation adheres to the laws you defined up-front — which it does not if it modified the laws.

No secret sauce is needed to be valuable

Ah, you're right. My mistake.

I'd like to have a model that just "knows" our code base and past issues, instead of having to query them for the same reason I don't want the model to have to query a dictionary to speak proper English.

I could be mistaken here, but I'm curious to see if it would improve both output speed and quality. Currently, every new prompt I start it basically spends 10-15 minutes "getting familiar with the code" which is both annoying to wait for and wasteful money-wise.

> We do the finetuning only on small semantic data were it helps a lot.

This sounds interesting. Would you care to expand a bit on how you do this? What is this semantic data?

> There are companies though which ahve this exact problem with programming languages you normally don't see.

This applies to my company. We have a substantial code base in our own dialect of APL. You think post-training would help substantially here?

Somewhat off topic, but I've started wondering if we can actually make local LLMs feel smarter than the frontier closed models, by post-training it for your specific use case.

Say Company X has a software product which consists of a million lines of code, including a ticket for every bug and new feature for this piece of software. Then wouldn't it make sense to try using an open weight model but post-train it on that specific code base while using the tickets to teach the model about past bugs and features. Not sure exactly how, but it could involve doing reinforcement learning solving a past bug on that historic version of the code base, and rewarding the model if it comes up with the correct solution (as defined by the linked PR which fixed the bug).

So essentially post-training your local open-weight model using reinforcement Learning with Verifiable Rewards (RLVR) on your software products history of bug reports and their ultimate solution. And the same for new features.

Surely it would depend on your model, since they need to etch this into silicon.

Comment threads like these are why I read HN comments.

Thank you for the insights, both of you.

> I still wonder why some of our key universities in the US aren't building Open Weight Models.

I'm guessing the $100m price tag plays a role. Plus the fact that it will be outdated next quarter.

> I think the opposite: I think the frontier labs have good margins on their inference unit costs.

Interesting. Good enough to make up for training costs?

Also, where can I read more about this?

> Good guess. The actual mechanism [...]

Why couldn't both these factors contribute to the result of the study?

> enforce right to repair

How much would this change Niño surface temperatures, you reckon?

It's a proof of concept that it's possible to etch a neural net into a chip and get massive performance (and efficiency) boost

If performance per watt is 100x better than GPUs (as GP link claims) then I don't think it's a hard sell at all. That's actually a cost reduction that matters.