Kostchei

interestingly, having actually done the law enforcement side of these investigations, 50% of them are local. And I understand that this is not 100% solution, but neither is any form of law enforcement, but that doesn't mean we should fail to attempt it.

Kids from the local uni having a lark, stalkers, vindictive ex employees, local gangs, criminals who understand their victims because they hail from the same community. These are your local hackers. Sift them from the nation states and international crime groups, then deal with the International as a matter of diplomacy. Because we do this so poorly locally, we have little ammunition to when it comes to diplomacy. "reduce attacks by your crime groups and we buy your natural gas, seel you wheat etc"

Want more motivation?- 75% of the local attacks by volume send funds back to terrorist or separatist organizations.

It is not an in-soluble problem. Sentences are a fraction of the answer, effective and receptive reporting processes are more important, then government backing for investigation and enforcement, then policy around home-team activities (ie don't do the bad things yourselves Mr Gov). Deterrence comes after all that.

In case you missed it, json is less likely to be written over than markdown by agents- something to do with the structure being more rigid

My favorite solution is a lower parameter 5 layer model trained on the data that acts as a local compression and response, a neurocortext layer wrapped around any large persistent data you have to interact with and ...... maybe also a specialist tool that spins up which is built with that data in mind but is deterministic in it's approach- sort of a just-in-time index or adaptive indexing

yeh- what is interesting is that it is way more viral and ... complicit than any of the doomer threads. If it does build a self-sustaining hivemind across whatsapp and xitter.. it will be entirely self inflicted by people enjoying the "Jackass" level/ lack of security

yeh the repo is Gas Town

These operate in parallel. Maybe you SDLC does that, the effort of each human developer sitting in a planning meeting, getting jira tickets, doing individual code (or pair or whatever), reporting back in standup, coordinating the next step, getting it QA'ed...

Yes, if your shop is well developed these work (10% of the time every time), but this is a structure to kick that all in to gear, as a repo, where all you need to add is unlimited machine cognitive power/tokens.

Maybe you need to add these gas town personalities to various parts of the existing SDLC, .....but..... you still need to track what they do and how- and you need them to intermediate between each other at 2am when they hit an impasse. Something very rare in most human cognition shops.

And word from the experimenters is.. it sort of works. Which is on par with most human shops. IMO. I don't have the money to burn to test at the scale Yegge is, but the small scale stuff I have done in this direction, this seems plausible.

Are you saying that people can't work out what to code using these? Or that code is not a worthy subject to use AI for? 'cause I got news for you... 1. Improving coding improved reasoning in the models. Having a verifiable answer that is not a single thing is a good training test. 2. Software has been used for fairly serious things. We used to have skyscrapers of people doing manual math. Now we have campuses of people doing manual code. You might argue that nobody would trust AI to write code when it matters. History tells us that if that is ever true, it will pass. 3. We are not going to run out of planet. It just feels to folks that there is not enough planet for their dreams and we get population panic, energy panic etc. There is a huge fusion reactor conveniently holding us in it's gravity well and spewing out many orders of magnitude more energy than we can currently use. Chill.

I think at Gas Country levels we will need better networking systems. Maybe that backbone Nvidia just built....

try the tools. Really. If you are remotely interested in tech or AI, try the tools Copilot this is not. You may be trolling of course. There are huge steps between these various tools, if you try them, for a smidge of investment, it will become obvious what the trajectory is.

It is like saying "I don't handwrite anything, I care too much about line spacing, I only use a dot matrix printer" when some one is trying to sell you a calligraphy pen and coloured inks, and you have only tried a ballpoint pen. You might be the wrong market, but they are not even close in use case and application.

(spelling)

so what you are saying is that for production we should use AI, and hand code for hobby, got it. Lemme log back into the vpn and set the agents on the Enterprise monorepo /jk

Yeh, I agree with this. My art (painting and building) comes at a much faster rate when I am content. Having time and metal space to contemplate colour scheme, being confident to start something bold: that doesn't happen if I am tired, preoccupied or depressed

The problem with laws that both the enforcer and the subject (enforcee?) agree are bad, is that enforcement is variable. And that leads to corruption. Every damn time.

Human metrics of intelligence have always felt like rubbish. We never did this well. I would describe intelligence as effective adaption leading to survival and growth or prospering. Memorization, comprehension, speed of response etc. those are magnifying factors that are valued, we view them as components of intelligence, but llms are proving this is not the whole, without effective application, they are not intelligence. Perhaps learning is the difference? How to measure that?

Someone describing string theory is the literary equivalent of fractal structures in snowflakes. Lovely, complex, possibly unique, but not proof of a level of intelligence- for the string theorist maybe it is intelligent, perhaps persuading someone to fund their grant, which enables them to eat, shelter etc. Might be a bit harsh on string theory. Saying it is proof of an amount of intelligence leads us to falsifiable statements.

We have 20+ services in prod that use llms. So I have 50k (or more) per service per day of data to evaluate. The question is- do people actually evaluate properly.

And how do you do an apples to apples evaluation of such squishy services?

damn, i was looking for army painter and citadel ranges :p

Except that a person born has rights granted to them about land and property mostly decided on capital. Very few farmers can refuse to sell the a megacorp. If you think this has no impact, find some un-used land that you have not inherited or purchased using capital and see how long you can grow food on it before someone objects. They might say it is a park, or a front yard, or a military training ground, but realistically, all those concepts of ownership were originally tied to capital and without substantial capital, you are a serf. A lesser. There is no commonhold for you to use. There are no unclaimed natural resources.

remains recoverable... for less than a training run of compute .It's a lot, but it is doable

I disagree. I have played since red box, played all the versions including 4th, I play 5th ed a lot.

For 5e.. 1st campaign, no houserules, 3 years run time including transition to online during covid 2nd campaign, no houserules, just a bit of re-skin warforged are necrons, right? etc in person 3rd campaign, added legacy items inperson, different group concurrent with campaign 4th campaign, some house rules on spellcasting (provokes attacks of opportunity etc), accelerated progression till 9th level, expand legacies to things other than items

Play Pirateborg, Starwars or Dark Heresy etc, don't play a 2nd campaign, no house rules. You just live with the short comings of the system, you won't be there for long.

That said, I ran a 10 year rolemaster campaign, so maybe I'm an outlier. But people who read the books and don't play a heap, seem to have a lot of strong opinions. "City X is a terrible place to live- says some tourist who has transited in airport".

twist, 30 years later the aliens tunnel up from under the concrete

<this is the way>

Depends on your life experiences and working environment. If you have worked in prisons and places with a lot of physical violence you can (some don't) acquire a distinct and accurate sense for emotion and threat, based on sound and body language. The actual words don't matter so much, but the interaction of tone, distance, stance etc, they tell you a huge amount. People can be saying "no" and be just asking a question or pleading their case, and they can be saying "yes" and mean "i want to kill you". I used to follow the tone, and when it was going to end badly, make sure I was standing behind the person who was about to start violence (being responsible for physical security in that environment), just as it was about to kick off... Pretty grim work. But yes, you can use your intelligence to learn that stuff. Don't need to be a puppy.

So the things I have seen in generative AI art lead me to believe there is more complexity than that. Ask it do a scifi scene inspired by Giger but in the style of Van Gough. Pick 3 concepts and mash them together and see what it does. You get novel results. That is easy to undert5stand because it is visual.

Language is harder to parse in that way. But I have asked for Haiku about cybersecurity, work place health and safety documents in Shakespearean sonnet style etc. Some of the results are amazing.

I think actual real creativity in art, as opposed to incremental change or combinations of existing ideas, is rare. Very rare. Look at style development in the history of art over time. A lot of standing on the shoulders of others. And I think science and reasoning are the same. And that's what we see in the llms, for language use.

In that it created a circuit inside the shoggoth where it translates between hex and letters, sure, but this is not a straight lookup, it's not like a table, any more than that I know "FF " is 255. This is not stochastic pattern matching any more than my ability to look at a raw hex and see structures, ntfs File records and the like (yes, I'm weird, I've spent 15 years in forensics)- in the same way that you might know some French and have a good guess at a sentence if your English and Italian is fluent.

I hope they learn from the terrible job of stewardship of tech the west has done and don't repeat our mistakes.

reminds me of every other tariff based failure in the past...

yeh. I see Dario saying "let's protect the US more" for no reason other than bias and "of course they improved over time" which feels like a mighty strong strain of copium. Very disappointing for a leader of an organization i respected. Assuming he speaks for Anthropic, and it seems he does, Past tense.

16 years+ for $200mn minus costs. Still pretty good, grats :)

You can get standalone/isolated versions of chatGPT, if your org is large enough, in partnership with OpenAI. And others. They run on the same infra but in accounts you set up, cost the same, but you have visibility on the compute, and control of data exfil - ie is there is none.

I did fingerprints from a digital photo in 2012 maybe earlier. Old mate was holding up drugs to be photographed. Bit of contrast, blew it up, sent it to fingerprint bureau and what would you know, we had those prints on file. Not the crime of the century and not absolute proof, but a damn good start for a case from a simple post on socials. More useful than most intel/hearsay that ends up in crimestoppers or similar channels.

Using the real world- as in vision, 3d orientation, physical sensors and building training regimes that augment the language models to be multidimensional and check that perception, that is the next step.

And there is very little shortage of data and experience in the actual world, as opposed to just the text internet. Can the current AI companies pivot to that? Or do you need to be worldlabs, or v2 of worldlabs?

Developer. Prompt Engineer. Philosopher-Builder. (mostly) not programmer.

The code part will get smaller and smaller for most folks. Some frameworks or bare-metal people or intense heavy-lifters will still do manual code or pair-programming where half the pair is an agentic AI with super-human knowledge of your org's code base.

But this will be a layer of abstraction for most people who build software. And as someone who hates rote learning, I'm here for it. IMO.

Unfortunately (?) I think the 10-20-50? years of development experience you might bring to bear on the problems can be superseded by an LLM finetuned on stackoverflow, github etc once judgement and haystack are truly nailed. Because it can have all that knowledge you have accumulated, and soaked into a semi-conscious instinct that you use so well you aren't even aware of it except that it works. It can have that a million times over. Actually. Which is both amazing and terrifying. Currently this isn't obvious because it's accuracy /judgement to learn all those life-of-a-dev lessons is almost non-existent. Currently. But it will happen. That is copilot's future. It's raison d'être.

I would argue what it will never have however, simply by function of the size of training runs is unique functional drive and vision. If you wanted a "Steve Jobs" AI you would have to build it. And if you gave it instructions to make a prompt/framework to build a "Jobs" it would just be an imitation, rather than a new unique in-context version. That is the value a person has- their particular filter, their passion and personal framework. Someone who doesn't have any of those things, they had better be hoping for UBI and charity. Or go live a simple life, outside the rat race.

bows