artrockalter

The internet became a single point of failure for all of civilization years ago and we're only realizing it now.

Part of the problem is that from that perspective there was no independent analysis done that could vindicate them. METR is absolutely part of the EA/LessWrong/rationalist ecosystem, so of course their investigation would validate that group’s arguments.

This stacks with the 50% discount in OpenRouter, making it $2/$10. https://openrouter.ai/openai/gpt-5.6-sol

In my experience LLMs do find a lot of embarrassing bugs as Linus says but it can constantly turn into a game of whack-a-mole where most of the bugs it finds were written in previous LLM sessions. It's a huge struggle to get it to actually fix the root cause of the bug instead of patching the symptom.

I know the company I consult for (not cybersecurity) is not in these programs and if attacked would need to use open weight models.

The Hugging Face incident is a great example of why open source models with defensive cyber capabilities are needed. Hugging Face did not have access to cyber-capable frontier models and kept hitting safeguards. Only by using the open source GLM-5.2 were they able to survive an attack. A world where open source models are banned is one where cybersecurity is impossible if you're not on OpenAI or Anthropic's allowlist.

The null is surely different in this case because we know that supplemented creatine goes to the brain and the brain uses it, which is definitely not a claim all supplements can make.

The fact that there's a biologically plausible mechanism plus mixed statistical evidence makes "maybe" a perfectly reasonable conclusion.

Is this methodologically similar to Anthropic's recent J-Space paper?

The task could be verifiable in the environment so limiting its CPU and RAM could be to discourage brute forcing the answer.

Both are true! The Confederacy did secede largely to preserve slavery, but the war was started to bring the Confederacy back into the Union, initially without the goal of also immediately abolishing slavery.

LLMs are good at writing complex regex, from my experience

When I use ChatGPT for work it frequently reads my Slack DMs even if they’re not directly relevant, so I’d question a lot of the premises of the article.

“they’ve been folded into Frankenstein positions that demand constant multitasking, social performance, sensory endurance, and emotional labor on top of technical skill”

My sense is that this is due to automation, not “neoliberal capitalism” as the author says. It’s much easier to automate a job if it’s a single task that’s done in a deterministic way.

It would be interesting if most of our confusion with quantum mechanics came from treating probabilities as independent when they are actually highly correlated. I don’t really know any physics, but I’m familiar with probability and this type of problem seems to be the most common error in interpreting probabilities.

SubredditSimulator was a markov chain I think, the more advanced version was https://reddit.com/r/SubSimulatorGPT2

It's so easy to ship completely broken AI features because you can't really unit test them and unit tests have been the main standard for whether code is working for a long time now.

The most successful AI companies (OpenAI, Anthropic, Cursor) are all dogfooding their products as far as I can tell, and I don't really see any other reliable way to make sure the AI feature you ship actually works.

Happy smart people are generally very focused and only care about a few things. Unhappy smart people are constantly getting "nerd-sniped" into focusing their intelligence on things that don't make them happy.

> It’s the same reason why most of the people who pass your leetcode tests don’t actually know how to build anything real. They are taught to the test not taught to reality.

True, and "Agentic Workflows" are now playing the same role as "Agile" in that both take the idea that if you have many people/LLMs that can solve toy problems but not real ones then you can still succeed by breaking down the real problems into toy problems and assigning them out.

I don't like using these KYC companies that I've never heard of either but I'd guess it's going to become a lot more common (especially for API use) as the models get more powerful.

An answer to the productivity paradox (https://en.m.wikipedia.org/wiki/Productivity_paradox) could be that increased technology causes increased complexity of systems, offsetting efficiency gains from the technology itself.

- AI wrapper to summarize long text files (sort of like the LLM-plays-pokemon agentic summary of conversation history)

- single page site for keeping track of what you're reading/watching (building for my parents who use pen and paper for this)