If you work at the edge of context you know what I mean. Even within context, if the system was truly intelligent, the way that Euclid was intelligent, why do I need /superpowers and 50 cycles to get a certain implementation right?

Why is the AI not one-shotting obscure but simple business logic cases with optimal code? Whoops pattern never seen before! There is no thought to it, zero. The LLM is just shotgunning token prediction and context management until something sticks. The amount of complexity you get out of language is certainly fascinating and surprising at times but it's not intelligence - maybe part of it?

Sell it as skills or whatever, but all you do every day is fancy ways of context management to guardrail the token predictor algorithm into predicting the tokens that you want.

You can see the cracks happening quite fast actually and you can almost feel how trained patterns are regurgitated with some variance - without actually contextualizing and connecting things. More guardrailing like web sources or attachments just narrow down possible patterns but you never get the feeling that the bot understands. Your own prompting can also significantly affect opinions and outcomes no matter the factual reality.

Or the impact of smartphones in 2003? Sure smart phones were considered but not the entire app ecosystem and planetary behavioral adaptation.

Just look at this example of a gene regulation network: https://www.researchgate.net/figure/Different-visualizations...

Amazing work, as always. I love neal.fun

Edit: also good to know that paper airplanes have officially beat the SR-71, F-104 or X-43B with altitude record.

It's obvious why the ultra-rich are building bunkers and hide-outs. Those are of course scams by the building companies, as they give a false sense of security, but the idea of what is REALLY going on is obviously out there.

I wanted to use NotebookLM as a tool to ask back and forth when I was trying to understand stuff. It got the answer 90% right but also added a random format, sounding highly confident as if I asked the spec authors themselves.

It was easy to check the specs when I became suspicious and now my trust, even in "grounded" LLMs, is completely eroded when it comes to knowledge and facts.

It's entirely based on the perception that LLM training & inference is here to stay at ever growing scales when the shortcomings of Artificial Dreaming are increasingly scrutinized. Not all businesses want to end up paying refunds to their clients like Deloitte [1] because the LLM hallucinated crap into their reports (and they failed to correct it).

[1] https://www.theguardian.com/australia-news/2025/oct/06/deloi...

Mandated to be accessible to EU citizens by 2027 when all Member States have developed a Wallet solution.

Not associated but learned through it at work recently, just awesome project and thought I'd share in this context.

[1] https://commission.europa.eu/strategy-and-policy/priorities-...

[2] https://eu-digital-identity-wallet.github.io/eudi-doc-archit...

[3] https://github.com/openwallet-foundation/credo-ts

The immune system is highly highly complicated and directed by huge networks of genes and molecules all up- and downregulating each other depending on internal and external factors. If things go "off balance" in this system the consequences could be dire.

You dont want firefighters hosing down your house from the inside when there is no fire anymore either.

The sketchy part is that LLMs are super good at faking confidence and expertise all while randomly injected subtle but critical hallucinations. This ruins basically all significant output. Double-checking and babysitting the results is a huge time and energy sink. Human post-processing negates nearly all benefits.

Its not like there is zero benefit to it, but I am genuinely curious how you get consistently correct output for a "complicated subject matter like insurance".