Quarrelsome

Having watched almost everyone in my career prefer to do what they're told instead of pushing back in order to build the right thing, I kinda agree.

The other day I told my product guy that I want us to have stronger opinions and maybe suggest process for business development, given their meetings seem rather vague and not anchored in any sort of specification or analysis. He looked at me like I'd just shot his son.

Commodities. They power the economy and they're worldwide. Ofc you want the free flow of goods. From the US's perspective who cares who runs what nation, as long as its not hostile and its stable then the commodities can flow from it. Then manufacturers can make whatever they want to create products to generate growth and wealth, the vast majority of which flows into the US and into their stock market.

As it stands the war in Iran is creating problematic issues with many petrochemicals and their derivative products making it harder for countries to generate wealth to invest in the US stock market.

> People with that level of power — because it is power — will do what they will. “The strong do what they can, and the weak suffer what they must”.

agreed and we should hate them for it.

> disagree with his politics

Its the character. Buying a global media platform to "promote free speech" and then shunt the entire thing and himself behind a particular political party. This guy uses his money and his twitter account to support neo-nazi hate groups in my country (Britain First, Patriotic Alternative, Hundred Handers, Tommy Tennames). He has no self-restraint. Its abhorrent. Its not "about politics" when the politics is about racial superiority and ethnic cleansing.

> Imagine if you couldn't post a thread about Microsoft products on HN without republicans feeling the need to jump in to talk about how much they hate Bill Gates for his donations to Kamala Harris's campaign in 2024. It's really tiresome.

and if Bill Gate was literally campaigning for the Dems, getting up on stage, handing voters cheques and THEN being employed by the government to run a department then you'd have a point.

He repeatedly crosses the line of what is appropriate. People with that level of wealth should stay the fuck away from politics and respect democracy instead of using their money and business to have excessive influence on democratic outcomes. Its unseemly and disrespects the idea of everyone getting the same amount of votes. I respect that donations have similarities but there's a difference between making a donation and being excessively public.

its not about virtue and its not even about starving. Its about making a global economy more stable so the world is richer and the US almost always profits from that.

he directed government money from strategic projects that makes the world more stable to ensure the US could waste it all fighting another dumb war. Its hard to justify, easy to criticise.

its cos he's a prick. The way he pretended to "fairly" sack devs at twitter when he was clearly just trying to cut costs. That he bought twitter to "defend free speech" and then started donating to the Republication party and promoting a very specific political position online. As a euro the Nazi salute alone is enough for me.

The qualifier matters, that guy has serious issues.

what people consider to be "intelligence" isn't about doing l33t maths, its about self-doubt. This enables us to mitigate against our cognitive biases and instinctual impulses and accept truth, which leads to more useful outcomes.

All that buying a calculator does, is allow you to do bad sums more quickly.

sure, but now if you're running an entire agentic run for an org, what's to stop you just replacing the org? Previously you didn't have the wadge to pay for all the devs.

if you're stupid then this technology just amplifies your stupidity which means you can fail faster. I also think the inverse is true.

It might be copium, but the only guaranteed thing that happens is change.

Remember its coming up millenilal time now and i have hope in this gen.

Tide goes in, tide goes out. Im low key of the opinion that these tools enhance existing attributes while displacing the advantages of unearned social or financial power.

So in the short term the tide might go out but then it might tsunami back.

Same goes for this terrifying political feeling around the world. Might simply be a boomer swansong and the reaction might snap back.

didnt the article say thay engineers still review change requests? They just switched from reviewing code to reviewing intent.

So "no direct human supervision" seems like a misunderstanding. Article even bigs up the purpose of engineers with a hilarious (to me at least) line about auto adversial reviews resulting in over engineered designs.

I mean you could have asked it to build a log classifier instead so next time it happens you either get the solution immediately or it autos into an llm giving you its answer.

You're wasting time and tokens here.

Apparently using sqlite aint learning.

DynamoDb has a max item size of 400kb.

That was months ago. It often surfaces pertinent info which is memorable cos it relates to whatchu doin.

Hear, hear. We had the entirety of human knowledge available online for decades now. Are people more educated? Nah, cos most people aren't interested in looking for much beyond validation.

Perhaps me finding your comment and replying has similarities but at least this is validation of curiosity.

I love llms cos history isnt typically indexed by the sort of questions i wanna ask but an llm can cut a wide array of sources by an arbitrary metric and i can keep poking for more and more.

I see a wikipedia parallel. There's peeps among us that will 2am doom scroll wikipedia, peeps that will only do that same on social media and peeps who do a bit of both.

give it a full spec of what you want. Full waterfall. You can then take it into an agentic session and get a lot out of it.

> Have you ever remarked something supply chain worries related in a system / initial prompt, then saw your agent being indefinitely paranoid about artifact signatures and hashes? Same idea. Float the idea of mistakes, it will invite the agent to run linters and maybe spin up critic subagents. Float the idea of only ever taking low-stakes hedges, it will frequently sound how confident it is in any given action, and may prompt for your approval more often. These are memes, yes, but they're not baseless. Not sure why you seemingly think they are.

Exactly, and that's the level of conversation people should be having, instead of giving it "don't fucking guess" and calling it a day.

Its just too ambiguous a construct to lead to reliable output and its an operator error.

> because I do not find this instruction ambiguous

YOU'RE HUMAN?!?!

Please explain to me, how an LLM can easily distinguish between information it knows and a guess.

Perhaps you meant: "within the available documentation for the technology that you're working on and its current version, after having also double checked the outstanding issues via the github repo. If the technology has widespread use then forums might also contain relevant information where users might share further information about known issues."

But that's a teensy bit more than "never fucking guess", which is the lazy and stupid version of it which is what gets people in trouble. Do you not realise how this alternate version explicitly specifies the search space where "never fucking guess" specifies pretty much jack shit?

"never fucking guess" is akin to "don't make mistakes" in how its completely useless as an instruction.

it follows instructions and "never fucking guess" is an extremely ambiguous instruction that increases, as opposed to diminishes the search space. It will lead to unexpected outcomes.

yes you're bad at using it. Your mistake is presuming everyone else has that problem.

Try considering the input a search query and the output a list of results of which one is arbitrarily chosen. If you get your query so specific that the amount of possible results are small enough that all of them are good, then it works. Big longer starting context, hand-written, use the LLM to improve it by criticising it (but keep it mostly hand written), aim to do more in a corresponding YOLO session.

inanimate objects don't interpret the abuse as input and have it impact their task. If shouting at a car reduced its fuel efficiency maybe it wouldn't be socially acceptable. I agree that none of this stuff has "feelings" but prompting like this will lead to alignment issues. Given the control we give these agents, that's inherently problematic.

Be comprehensive, mostly positive and straight forward, especially in opening specs and base .mds. Using this tech is a responsibility because it can do a lot of stuff that you don't expect it to. Telling it "NEVER FUCKING GUESS", it going to increase that possibility because its an extremely strong and extremely ambiguous prompt.

Its entirely different because a comment isn't compiled. It's entirely skilled. Conversely, a prompt is part of the input to an LLM while contributes to both its output and operation when agentic. Its dangerous operation and will lead to alignment issues. Its part of why negative prompting outside of a one-shot is a bad idea due to "don't think about a duck" issues.

In terms of the statement "never fucking guess": How does an AI _know_ what a guess is and what a guess isn't? Write me that function without infinitely looping context. Its a disaster of a prompt which has an unhealthy attitude as well as a tragic anthropomorphisation of the model.

> Their mindset is healthier than yours, because they realize AI isn't alive.

You're wilfully missing my point. Its not about "the AI having feelings" its about the operator having issues and as a consequence building an alignment issue into their output and causing problems. If you stack these sorts of prompts headless in yolo mode into capable models then you're asking for trouble.

telling a model not to guess. How could it define what a guess is?

telling an AI to "never fucking guess" is stupid and abusive. Thinking its fine, is failing to add together the mindset that would generate such a prompt. Its a mindset of someone upset, frustrated and lashing out, asking for something that the model can never do. Its stupid and angry which are very much characteristics that are abusive.

humans have variance too. Its never zero under any technique.

The instability is due to variance. You form comprehensive inputs to constrain that variance. Thus it is no longer unstable.

Consider it similar to searching Google and arbitrarily selecting one of the possible answers. The less answers you get back, the less the variance. Write long and comprehensive queries!

There are still risks but they're all tractable as long as you don't get too comfortable and get lazy.

I think you need to go much, much, much deeper for that.

> Could you unearth that story?

Sure:

https://x.com/lifeofjer/status/2048103471019434248

His prompting is likely confusing the agent. "NEVER FUCKING GUESS!" is a terrible prompt, combine it with "NEVER run destructive/irreversible git commands (like push --force, hard reset, etc) unless the user explicitly requests them" and you can easily get issues. That "unless" is awful and the idea that a model can tell if its guessing or not is very non-trivial.

Just generally the language he uses and his attitude demonstrate that he's not the sort of person that should be doing this thing. He's not skilled enough to understand the risks.

> I don't see why giving an agent the access necessary to perform actions would be "unhealthy prompting".

Sorry, two different issues. The major factor in this happening is IMHO:

* Don't give agents the keys to production. Instead have them write tooling that you can test and you press the button in the tooling. No AI, deterministic process.

* Abusive prompting - this can create alignment issues because if you're angry then it might conflict with previous input, this can make a model unsure about what to do and act in unexpected ways. Also it creates the "don't think about a duck" problem.