eithed

Thank you, that clears it up that it's indeed still Google. Oh, it's even said in the article.

I'm confused - isn't a contractor (Tuike Finland) doing and being responsible for this work?

Don't think that I would be able to find information in the phone book whether given person is right leaning, or left; whether they said something racist or not, etc

The most hilarious moment for me was when my question that was 5y old was closed as a duplicate of one that was couple of months old. I provided an answer on how to work around the issue; the new question didn't have an answer, but it had more points. From what I read the original issue was still there. At that point I was rarely visiting the website either way, but the absurdity of the situation was staggering.

I'd think about it this way - with LLMs I'm still a human with one brain, one set of hands, one keyboard, one screen.

I can delegate more quickly, but still I'm constrained to what I can respond to because I can focus on limited amount of thing at a time. It's like I was given multiple sets of hands (so your centaur parable stands), but that doesn't make me faster if I need to do multiple things at the same time - have you tried juggling, chewing gum and skipping at the same time?

I don't know how people are handling constant context switching - ok, I can trigger 5 agents that work on different things. One comes back to me citing a decision I've made 30m ago; then another, citing a finding to something it mentioned 45m ago; then another, then another. In every one of these cases agents are unhelpful as all hell - they overcommunicate, lie, scope creep, expect you to pay attention to every single thing they mention in passing. You've created a plan? Fuck you - agent has exceeded the plan with 2 items. You ask about these two items? It's actually 8. Ok, what are these 8 items. Sorry, I was wrong, it was 10. And in the background there are 4 agents now waiting for your decision, where you have to familiarise yourself what they were doing and what were the decisions that you've made.

Now the gatekeeper is saying "I don't want to be a gatekeeper anymore for this one particular gate, because people were mean to me. I'll however continue being gatekeeper for these other gates that drive my business either way."

Heartbleed

Your harness tells agent to do A when it encounters B. When agent encounters C and it doesnt know what to do it does D and you correct it to do E. Harness keeps track of this and makes agent do E next time it encounters C.

That's a very simplistic view. I'm very good at writing tests and I'm happy to offload writing tests to LLMs, as long as they follow the structure I've provided.

You're saying that I must not be good in writing tests, irregardless on how I feel about test writing.

no. weather isn't intelligent, it's not alive, yet humans treated it as a god in the past and gave it agency; we understand how it works, yet cannot control it or predict it.

but I digress - the main thing: these systems are not alive. They pretend to be.

I like this - it reminds me that there are systems (ie weather, economic systems) that are in no way intelligent, but are emergent and react to interactions.

Don't think this is an offensive position and don't think you're telling people what they ought to like; like author you're expressing your opinion

It all comes full circle with chinese models being available free for all humanity to use

I think it was more tongue-in-cheek answer = not building torment nexus isn't an option! Also it's another narrative Americans are pursuing being the world police.

If you can quantify quality/reliability/understandability, you can tell LLM what kind of code do you expect. If not, you get whatever.

At my current place we not only have automated tests, static analysis and static rector (linting, but also automatic pattern matcher for problematic code) but also: - architecture tests that define relationships between application layers - ADRs that guide developers (and agents as well) that communicate how new code should be written and how existing code should be treated

I find that "how code should look like"/"what code should do" is an ambiguous idea that always is preached, but never defined = everyone's idea of quality is slightly different and only looking at existing code you tend to align. Everyone's idea of what the product does/should do is kept within their heads. If we define this knowledge in writing LLMs can not only write code according to the patterns that are thus defined, review existing code based on these documents, but also actually read acceptance criteria documents to check if the code does what it's intended to do (gherkin)

Same goes for understandability - if LLM applies one pattern this time, another pattern another time, if you have multiple coding patterns then that hurts clarity. Sometimes LLMs work as common denominator thus achieving clarity, but I find that actually giving LLMs reference works.

Can you quote your sources? The article doesn't mention anything about the regulation you're stating as a fact.

Same here - idk why are people jumping on the wagon of attributing the restriction to EU regulation. Looks to be a US compliance banning social topics that Trump administration is pursuing (but again, idk if that's the case, this is guesswork on my part)

Where did you get this from? I don't see anything in the article mentioning that specific regulation

ShareYourAISetup folks create a skill that user can run locally that collects information re their setup, formats it and sends to dedicated endpoint. Result = fairly uniform reports where people share what they're comfortable sharing.

No requirement from given user to spend their time. Users see what the skill does.

I guess atm they dont know what questions they want to be asking

What even is this shit? Every time I interact with models they do that, or any other variation of "let me make decisions on my own just to get the task done" - is all of this misalignment now? The most egregious to me was when model asked itself if it should proceed with dangerous command, gave itself approval and then wiped my local DB.

What can I even do with this report? "Our models don't follow what users ask them to do", no shit sherlock.

Right, so really any fresh file is more or less fine (would still evaluate cyclomatic complexity). It's the quantity/frequency of changes that matters (or CRAP)

> If the change adds a complex function to an empty file, it's likely just a complex single problem.

How do you determine that this is correct = adding complexity to a single file vs adding less complexity to multiple files that this one file then orchestrates.

That's why I tell my Claude to keep its responses short :V

Then why wouldn't they use a bulletin board that is dead? And according to https://collusion.wiki/ OpenAI did see the results on this board on 21 Jul, so why not make agents delete their traces?

I'm dubious - if the agents were so smart that they've used a message board to coordinate and if they were to do it on other message boards that were not found, then why would this one be found? What makes it so different?

Is it just me, or is it advertising? "Look at how smart our models are, they used this website to coordinate and share guidelines!"

I can see porn on x.com without any issues (this is from top of my head regarding widely known websites that show adult content without any checks)

On the flipside, I cannot buy any adult orientated games on Steam, because I don't have a credit card

Dependable.

I love that Gibsons cyberpunk is "realistic" and mundane (at this point): a girl replaces a virtual singer (Mona Lisa Overdrive), another girl steals glasses containing plans for urban redevelopment (Virtual Light), virtual world is as inhabited as real one and you only need helmet to access it (Pattern Recognition). We already are living with these stories and dystopia is already here, to a degree.