Kostchei

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 years of development experience you might bring to bear on the problems can be superseded by an LLM finetuned on stackoverflow, github etc once judgement and haystack are truly nailed.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

I mean fkcu me when they have those things, however, maybe they are just lazy and their judgement is fine, for a lazy intelligence. Inner-self thinks "why are these bastards asking me to do this? ". I doubt that is actually happening, but now, .. prove it isn't.

OP runs 300 essentially free forums. It's just too much.

<insert manic laughter> I mean that might happen, but why get funding? Why not move to market immediately? Without debt. Get in the thick of it. Just do. You want a wage or do you want a product that does the thing. Because sometimes, with the llm, you can just build the thing. The marketing, the compliance, you might hire for that, or you might also out-source to the llm.

Why would you hire? Either it works- in the sense of does the job and is cost effective- or it is not.

Is there a situation where paying 100's of k of wages makes a thing suddenly a good idea? I have doubts.

This my technique- test the output, make sure it works (in the sense of outputs) the way I want. Test the input edge cases, move on. Occasionally when I can't get it to do what I want the llm suggests things like logging of output between functions etc- in which case they get added, but at the end I ask it to take out all the logging and make the code more concise.

And sometimes it breaks in ways I can't fix - so rolling back or picking a new patch from a know break point becomes important.

16 hours for my first azure pipeline, auto-updates from code to prod, static app including setting up git, vscode, node, azure creds etc. I chose a stack I have never seen at work (mostly see AWS) and I am not a coder. Last code was Pascal in the 1980s.

3rd app took 4 hours.

Built things I have wanted for 30 years.

But yes- no code understanding, brute force.

verisimilitude, not coherency. The real world doesn't appear coherent, why would a fantasy one. The consistent application of style and rules- that gives verisimilitude, and that's what counts.

They (mega) may have started to do this, but for several years I used to come across saved text files with lists of of mega links on most pedophiles computers. A different unit in the police sometimes checked the links and anecdotally they still (months after seizure) were valid and contained Child Exploitation Material (CEM). Often the same material stored on the device that I was wading thru. (insert cannot unsee meme) I am not talking about fakes, I'm talking about photographic documentation of the worst abuse held by offenders either as memories or currency. So when Mega came under fire I didn't really care that if came from the "copyright" end. I understand that Mega faced a large task to address this. Not simple. But the fact that CEM was hard for them to sort from the large corpus of "pirated" material seemed like a feature, not a bug. A least from a site design perspective. Selfishly, my sympathy is low in this case. Motivated by my own secondary exposure to some horrible shit as a result of the legal process.

not to defend its "you must accept updates" insane /inane fail, but, the suite of crowdstrike inc falcon stuff we have enables the response side of EDR pretty well, and for a mixed windows, linux, mac shop, where we would like the same agent on all systems, it does a better job than most. Not as good as Jamf on Mac mind you, but better than than most "windows ecosystem". And if you run jamf for policy and detection, but not response, you sort of get it all. So, that's why not "just defender" - at 10k+ systems the anti-malware is just the beginning. What do you do when that fails and ...yeh.. anyway.. there is more to it.

As to why windows is not more locked down- that's on the shoulders of the admins. But out of the box, you are right, it is to permissive. But apparently users and management like it that way.

As a user of Wiz (and I like what I get from them) ,I am relieved . 1. huge valuations are bad for customers- that investor cash has to come back from somewhere 2. Google has a habit of subsuming products into their stack and either sunsetting them or holding them so close that nobody uses them (beyondcorp etc) *spelling

This is the way. Robotics, self driving cars, mobile devices themselves, either phone or glasses.... They can generate the data needed for the next steps. (pun intentional) Text was easy, the process has been defined. Multi-modality improves the models. The next set of data is the world. Everything we currently observe using our senses and data outside our senses like radio waves. Build a better world model.

he regularly does- and gets various threats from bad and dubious actors as a result. He and "have I been pwned" (troy hunt) are genuine positive actors with IRL stakes in the game.

We have run that test.- generate random string(not by llm) names of values- ask the llm to do math (algebra) using those strings. Tests logic, 100% not in the data set GPT2 was like 50% accurate, now we up around the 90%.

a generation of consultants is already doing that- look at the ruckus around PWC etc in Australia. Hell, look at the folks supposedly doing diligence on Enron. This is not new. People lie, fib and prevaricate. The fact the machines trained on our actions do the same thing should not come as a shock. If anything it strikes me as the uncanny valley of truthiness.

hmm. I ended up with a few 750+ line chunks of js, beyond the ability of chatgpt to parse back at me. So my go-to technique now is to break it into smaller chunks and make them files in a folder structure, rather than existing inside a single *.js So readability is an issue for me- even more so because I rely on ChatGPT to parse it back- sometimes I understand the code, but usually I need the llm to confirm my understanding. I'm not sure if this scales for teams. My work has sourcegraph, which should assist with codebases. So far it hasn't been particularly useful- I can use it to find specific vulnerable libraries, keys in code etc, but that is just search.

What I really need is things like "show me the complete chain of code for this particular user activity in the app and highlight tokens used in authentication" ... - something senior engineers struggle to pull from our hundreds of services and huge pile of code. And so far sourcegraph and lightstep are incapable of doing that job. Maybe with better RAG or infinite context length or some other improvement there will be that tool. But currently the combined output of 1000's of engineers over years almost un-navigable. Some of that code might be crisp, some of it is definitely of llm-like quality (in a bad way)- I know this because I hear people's explanation of said code and how they misremembered it's function during post mortems. Folks copy and pasting outdated example code from the wiki etc. ie making things they don't understand. I presume that used to happen from stackoverflow too. Engineers moving to llm won't make too much difference IMO.

I agree, your points are valid, but I see "prompt engineering" as democratization of the ability to code. Previously this was all out of reach for me, behind a wall of memorization of language and syntax that I touched in the Pascal era and never crossed. 12 hours to build my first node.js app that did something in exactly the way I had wanted for 30 years. (including installing git and vscode on windows- see, now I am truly one to be reviled)

I speak English, not any code languages.

I recently learned javascript this way - knowing English and other languages (Thai, Mandarin), I kept pasting my pseudo-code, asking GPT to convert it into javascript, and then iterating on that. Took me a week or two to get the language without reaching for any sort of manual/tutorial. Speaking without experience - it would take me 2-4x as much time to learn the same thing otherwise.

:) My story is not entirely true, but close. My point being llms are learning language and logic (mostly English currently). Programming languages are just languages with logic (hopefully).

And if you think the ability to shorten meaning of a complex idea is exclusively the purview of code, think of a word like "tacky" or "verisimilitude"- complex ideas expressed in a shorter format, often with intended context with significant impact on the operations in the sentence around them.

This is my new capability- I am not a coder or a programmer and I can get things built, in my own time, at my own speed, solo.

Would it be better code if someone with 3 years of university and 5 years of coding practice did it? Yes, very probably, but the gap seems to be narrowing. Humorously I don't know enough about good code to tell you whether what I build with llms is good code. Sometime I build a function that feels magical, other times it seems like a fragile mess. But I don't know.

Do I know "javascript" or "python" or the theory of good coding practice? No, not currently. But I am building things. Things that I have personal, very specific requirements for. Where I don't have to liaise or berate someone else. Where I don't have to pay someone else. Where I don't share the recognition (if there is any ever) of the thing, I and only I have produced- (with chatGPT, Gemini and most recently llama3).

Folks have been feeling superior for 70 years and earning a good living because they spoke the intermediary language of compute engines. What makes them actually special NOW is computer science, the theory- the languages, we have very cheap (and in the case of open source local models free) translators for those now. And they can teach you some computer science as well, but that is still time and practice.

I'm the muggle. The blunt. And I'm loving this glowy lantern, this psi-rig.

sure, but if the phone company says there was an sms, when you open the phone it looks like there is an sms, and cellebrite extracts one, which you then present in a nice power point to the court... then your accusation as to tapering with the evidence and where in the chain of custody it occurred had better be good, or all you are going to do is annoy the court (as someone with dozens , maybe hundred+, of cellebrite reports in various court records) This is what armchair lawyers tend to forget, especially coders tend to forget. The law is not code. The judge is not a complier. When the law does not represent reality (this guys sells drugs and uses his phone to do it being a common reality) , then it becomes news-worthy.

But normally people just look at the reality and go "oh yes, this tool extracts the stuff on a phone and turns it into a pdf/html, how convenient". 99.99% of the time, the time a drug dealer alleging he has no knowledge of the 100's of deals on his phone is about as realistic as your 5 year old nephew with cake smeared on his face denying he ate the last bit of cake... and is treated as such. Should the act of selling drugs be a crime?- completely different topic.

Depends what the incentives are. Some folks like the community- that doesn't change. A website like honestwargamer or goonhammer has a clan. Mostly because they are insightful and friendly. But if an AI scraped their content, it would have to build a better goonhammer? Maybe. But I bet it would have half a dozen nerds contributing and tinkering at the backend, painting the new minis that came out, talking about tournament results etc.. That is current, constant fresh relevant content, based on IRL activity. Very hard to replicate. For honestwargamer... Build a better twitch stream... that seems even less likely. they run a round and stream games, review results, collate stats. And then engage with the audience in twitch.

So when you say "contribute to the internet", this is what I consume and.. I'm sure there are similar examples in every niche- fishing, golf, coding, AI art creation ...

No I don't see this as gloomy scenario, and content creators- the goonhammers and honestwargamers, creatives, are still going to get paid (a bit, they were never rich), maybe in new ways.

Most of the arguments I hear from my mates-who-code about adding deterministic old school things like buttons and "ui" on top of LLMs is "sometimes it's faster and more specific"... Ok, sure, but go to voice to text, and give the llm in a voice version, to a grandma. They can just use it.

and if you, the programmer need specifics "I want to do root cause analysis of these 20 incidents, break them up into mobile and web app, then draw common threads in each type of RCS".. then write/say a better prompt that exactly details what you want. Might take 2 or 3 goes, but you will get there. Adding metadata tabs like some vector-db franken-sharepoint seems like a step back. At least from a Rich Sutton world view. Let the LLM work it out- and if it fails, improve it for a many uses cases, not just one.

IMHO (tldr- strong disagree)

Maybe, but I think it is more likely I will try a different type of project, a different stack. See if that is the only easy path. Try something with graphics (a visual map) or that uses the llm api (generate a narrative etc). But my mate who is a programmer agrees with you- he sees the same thing- it is a good way to learn while being productive.

No different from "the team that built that is all gone, they left no doco, we assumed X, added the feature you wanted, but Y happened under load" , which happens a lot in companies pushing to market older than a minute.

My default assumption now, after watching dozens of post mortems, is that beyond a certain scale, nobody understands the code in prod. (edited added 2nd para)

With no experience in java, no coding for 30years since pascal in high school, no previous use of git or github, no hands on experience of the azure stack... I stood up 4 static web apps that do things I want in my hobby in 4 weeks- the first one took 12 hours including being shown git, installing npm etc etc. The last one took me 40 minutes. They do things for me in D&D that I have wanted for 20 years- now that capability is accessible. Whole monster manual ingested into a level, terrain and faction based encounter system that give ranges and features for the encounter ie a battle map. Scaling encounters suitable for the party at any level that theme with the terrain and dominant faction. The best thing about an MMO but for 5thed dnd.

Did I learn a bit of java and css and git?- sure, but I was up and running in about 4 hours with a mvp for my 1st one. There is NO way I could "learn" that in that timeframe. I just asked chatGPT 4 how to do it, and it told me. When I didn't know how to commit, it told me (actually I didn't even know the concept). It held my hand every step of the way.

I didn't need to learn something first, I just did it. And I have started doing it at work. "hmm 4 GB of fortinet logs in 20 files of gzip on mac.. how do I find a host name in that? - chatgpt.. oh- 1 line of zgrep.. never heard of it- hey it works.."

admittedly, I am bathed in tech, been hanging around folks talking about projects for years. But NOW I can execute- the problem? When it hits about 500 lines of java- maybe 10 functions, it is too big to drop into the prompt to debug and I don't know enough to fix myself. Solution, make smaller apps, get them working, create data files to reference in json, chain them together. eh, not perfect, but good enough for hobby.

Beware- fools like me who know nothing will be bringing code to production near you soon. Cool that you like to learn stuff, but syntax bores the crud out of me, each to their own, I'm just going to make. I find it more satisfying. Terrifying that code born like mine will end up in someone's prod, but it will.

Are you kidding? "easy to stop"? When LLMs are integrated into law enforcement, banking, research, education, logistics... All these areas have people building backend systems leveraging the current llm tech and pipelines to plug in the coming tech. If we reach a tipping point where these things become aware/act truly autonomously, what are the chances they do it before we notice? People are renowned for implementing things before understanding the consequences.

And what does charisma of AI alignment folks have to do with anything?