zbyforgotp

Schmidthuber has an answer: https://arxiv.org/abs/0812.4360

For a more practical approach you need to use proxies: https://zby.github.io/commonplace/articles/what-an-automated...

Here is my proposal (or actually our proposal - me and my agents): https://zby.github.io/commonplace/articles/what-an-automated...

On the flip side if there is a bug in one of my systems I would rather get a detailed bug report from an llm than a user message “xxx does not work”.

This is so obvious that in practice it doesn’t buy much, but everyone is still propagating that silly news. This is the real malware, a mind virus.

Too late for fixing it - but of course I meant https://www.qubes-os.org/

In the limit we want the llm to write the code (like in RLMs).

We don’t trust llm execution- so we add user approvals. But task decomposition calls for co-recursion between code and prompts. This means that the approvals should be evocable at any depth. I think we need some kind of protocol for that (à la the Cubes OS protocols for cut and paste between vms).

Maybe a workaround could be to use bubblewrap of the scripts ther recursively call the llm (and run the agent in yolo inside the wrap).

The idea is that the spec is somehow compressed in comparison to the code.

You need supporting environment on both sides of the sandbox.

Frontmatter should now be in the markdown standard.

I have been struggling with that. Thanks! Let me reword it - natural language lacks a strict semantics - so also programs for the llm machine ( I.e. prompts) cannot have it. LLMs always have to project from all possible semantics into one (are there any experiments with superpositions?)

What is the root cause of this? It is said that AMD is hardware company and neglects software - but recently they issued lots of declarations of becoming software firs now.

The point is for the agent to have an index of available skills so that it can decide autonomously if it needs to load one.

https://zzbbyy.substack.com/p/what-are-skills

In my experience it doesn’t work too well with codex, but I expect llm providers to train them on that use case and improve the situation soon.

At some point evolving software instead of designing it will work. Now the evolutionary pressure leads towards churning more tokens.

Property is a local low - it applies to a thing that exists in one place. Intellectual property is trying to apply similar rules to stuff that happen remotely - a text is not a thing, and controlling copying might work in some technological regimes while in others would require totalitarian control. When you extend these rules to cover not just copying of texts but also at the level of ideas it gets even worse.

There is something not right with expecting that artificial intelligence will have the same characteristics as human intelligence. (I am answering to the quote)

That’s why we need progressive hardening: https://github.com/zby/llm-do/blob/main/docs/concept_spec.md

The American labs are paranoid. The secrecy kills innovation. Open Source means ideas can meet and have sex and produce offsprings.

Destroying high trust society is a lot of fun.

TLDR: ai changes the economic calculus of software development. It makes automated testing more beneficial in comparison to costs.

I think he is right.

They should just give the user some control over this

Just to be precise. The sepsis alarm was not blocking x-ray or antibiotics but rather suggesting them.

“”” Agyare had instructed Banerjee to hydrate Sam right away but to wait for the results of Sam’s lab work before ordering a chest X-ray or the strong antibiotics used to treat sepsis. “””

The problem is that people gravitate towards more impersonal relationships themselves because it frees them from the complexity of social calculations. We escape small organisations, we try to be independent from each other and prefer to depend on impersonal institutions.

I think our minds don’t use novelty - but salience and it also might be easier to implement.

Zuck poaches AI devs and places them under Wang - how does that work? Wang doesn’t make impression of being a brilliant researcher or coder just a great deal maker (to put it diplomatically).

I started school in 1980 and I don’t remember this. Also books don’t use this typeface no matter how old.

There was no typeface change for “ż” - the other typeface is sometimes used now as it was 30 years ago. See the foto at Wikipedia: https://en.m.wikipedia.org/wiki/%C5%BB

They can do the milking without selling - don’t they?

A practical rephrasing of LSP: subclasses should be subtypes.

Ok, maybe that was exaggerated - there are differences, but they are shrinking.