So basically "Reflections on Trusting Trust"? Children of changed society are also changed?

Except that your view doesn't change the reality, and both views are true at the same time.

So you decide that the reality is another view and that's settled?

Look, I could be wrong. That said, I think AI is both in some conversations a tool and in some conversations it’s short hand for “Capitalistic Acceleration”. With each iteration of RSI, this second context is increasingly the more important mental model.

Maybe reread my comments in both contexts.

Capitalistic acceleration will most definitely keep trampling on anything in front of it. Because, the humans with capital will steer it that way.

Eg OpenAI steering the “AI the tool” at Navier Stokes on a rumor (inaccurate as we found out) of Anthropic having solved it.

Other busy people in real world are then subjected to your generated content. Why bother generating then. Just write down the same thing you prompted llm with

It all boils down to whether the product (code , documentation, whatever) is polished, easy to understand and works as advertised. Beyond that, I don't give a rats ass about whether the intelligence that created it was human or artificial

Oh, thanks mom! I will make sure to put on my rain coat before going out to play.

You need "dreaming" phase;large context high quality llm to prune and condense the memory.

But does it actually work, or is it something made up because it sounds cool.

It's an improvement, but obviously you're still limited with effective context size, and everything over it is marginal improvement.

I'm curious about this, where'd you learn that?

There were quite a few people from the AI community on X sharing screenshots.

What about general knowledge you can get out of it before hallucinations start?

I do not rely on any LLM of any size for general knowledge baked into the weights, they all hallucinate and that is the wrong way to hold them imo

I think there is some merit in that smaller models cannot memorize so much of the training data, i.e. that they are less likely to do copyright infringement, and by analogy not having memorized SDK / API surfaces that have since changed from the training data

> I do not rely on any LLM of any size for general knowledge baked into the weights

You have to rely on it to a certain level for agentic/coding work, presuming that's the general subject we're talking about here... For instance I recently encountered a project where it would have been a lot worse if the LLM didn't already know "what is" xterm.js and a bunch of its associated npm-related/node related software. If it was still smart but had to google and find results for everything it would have been a lot more time consuming and risked sending it down a wrong path.

for sure, there is a minimum size and knowledge base that is required to be useful

at the same time, search may find newer or better alternatives, and you can always specify specific technologies you want to use, I typically do this when starting a new project

Storing general knowledge in VRAM has always been a dumb idea in the first place.

It did OK on schlongbench v1.0 (test of a specific niche word that doesn't make it into smaller LLMs) but it sure does love to count words

https://pastes.io/r8F1AY8h

Qwen 3.8 27B beats Opus, Fable and GPT 5.6 by a comfortable margin on the AA-Omniscience Hallucination Rate benchmark.

Best way to apply?

ning.wang@deepseek.com

Cheap investment to boost ones CV.

You load it, tell it what's in and it sets the program, tells you what it set and why. You approve and off it goes

I do that today by turning a dial and pressing the start button.

Not saying it's the superior way, but to be honest I wouldn't know how to start some washing machines without consulting manual first, having very dumb simple appliance myself.