- cross-posted to:
- games@lemmy.world
- cross-posted to:
- games@lemmy.world
You must log in or register to comment.
Opened his mouth.
- broke the Intel fuckup, which turned into a huge story
- did a MASSIVE investigation into GPU smuggling into China, which turned into a huge story
- just recent did a deep dive into Jensen Huang, staying fairly pointedly away from the lurid scandal bullshit and diving much more specifically into the corruption and market manipulation angle
I say this with complete seriousness: I trust Steve Burke’s journalistic integrity and investigative incisiveness, and that if the team he’s put together, a damn sight more than I trust pretty much any “major” news outlet these days.
He and the guys at Gamers Nexus might be the best tech investigstive journalists in the world right now.
Apart from being a certifiably giant cunt, yes he is a lunatic psychopath.
That channel is dynamite.
Why are all the Sweenies so disappointing.
It’s in their jeans.
What none of these tech idiots get is this was never the scarcity anyone wanted to solve. We already had an avalanche of games. We didnt need genAI sprinkled everywhere to make it worse. Nobody would have cared if these were scoped for programming and dropped onto dev machines, but again programming was never a bottleneck. You can’t generate away out of the art, maybe you could have used it to generate bones and the shit ass parts of 3D modeling to free up artists to just create. It didnt have to be this way. But they wanted their centralized surveillance machines. None of this scale up is actually about genAI. Its about surveillance.
/rant, sorry
I would say they do put some value on the fascist mode of production for its own sake, and there’s definitely an anti-labor angle in there somewhere.
It didnt even have to be fascist either, if someone told programmers “hey, you want super auto complete? All it will cost you is stackoverflow” we would have rejoiced. A lot of us already wanted to share a lot anyway. But yeah, the way it was actually made was highly destructive and continues to be destructive. The only true way forward for this technology is local unless we want extinction. In the meantime it is fun seeing chip shipments turn into The Fast & The Furious scenes.
Ironically without a new stackoverflow to build new knowledge exchanges, human to human, LLMs have already peaked as programming assistants. Hence why they discourage anyone from moving past python 3.12 for instance despite the coming end to LTS. LLMs simply do not have the training data to engage with frontier programming properly.
I mean I dont know about you but the local ones do what you ask. They do feel like they have peaked and I would say its time to move on if someone is searching for intelligence. I myself have more positive programming interactions in person and on socials than I ever did on stack. But more to the point I dont think their place ever was frontier anything like people try to use them for. I think their strength is in reducing monotonous tasks we can define. Whether the user has the capacity to use it correctly is a tale as old as time and no different here even if the “tool” happens to be the most controversial thing made in modern history, with good reason.
Well put. My day job is training AI for major corporations and governments and my point has become the same: its an advanced autocorrect feature, built on a model and architecture that guesses the best answers based on language patterns and nothing more.
For programmers to use them to remind themselves of the fastest way to iterate a list, or map a matrix, it is amazing. But anything more than a single function in size and its immediately risky for the same reasons it is a good guesser.
Programmers have turned out to spend MORE time coding than before LLMs https://metr.org/blog/2025-07-10-early-2025-ai-experienced-os-dev-study/, from my experience that is due to not understanding the limitations of the architecture. But that so means admitting they aren’t that game changing as we were sold.
Especially with art. LLMs are absolutely dog shit at art if you need to iterate for accuracy.
it is structurally architecturally fascist.
Because they stole all the work to create it. If you setup a program to collect data and rewarded people for it, then just returned it to the people when it was made, it simply wouldnt be fascist and it would be more purposed instead of whatever generalist shit they are doing now.
Edit: computers and tech related to it can be however we want, its not architecturally anything unless we make it that way.
no. stealing is not structurally or architecturally fascist. try again.
Your own brain is scrambled by AI psychosis. I just said voluntary program but you can no longer read you only hate. That is the thing that will tighten our downward spiral and promise our downfall.
you like spirals don’t you.
And actually after watching the video and thinking, you dont even need AI for that. Classical computer programs do generate bones good enough for you just to edit them. You could solve it with just math to “free artists”
I work with a lot of software and am seeing this everywhere - problems that would be efficiently and cheaply solved with classical approaches but are solved with slow and expensive inference.
Just recently I saw an llm call to extract a postal code from a structured input. Regex would be trivial to solve it, but the person who set it up used fking LLM. What a joke.
but are solved with slow and expensive inference.
And also non-deterministic
Yeah I’ve run into this while playing with models/ agents in my homelab. Anything I’ve thought about using them for I’ve come to the conclusion that in every case I’ve got in my lab they can all be completed with shell scripts so why waste the resources?
There was recent drama around a completely vibe coded attempt at multiplayer Daggerfall. All of the prompts were leaked into the original github repo. One of the best was when Claude was asked to change an inventory icon for a fish. Claude’s response was to create from scratch a pixel art editor just to spit out a 36x24 PNG of a rotten fish.
And these guys wonder how they can possibly be burning through literally thousands of dollars in tokens every month 🙄
PEBKAC- Use codex for images, Claude for code.
BUT that is hilarious 🤣😆🤣
hell yeah, maximalist fascist mode of production!
I kinda wonder how they end up there. Couldn’t they have just asked the bot to make the regex code if they were already using it? I imagine I am just being fucked by my biases here cause I would already to know to ask it for that.
Also, the AI companies aren’t actually incentivized to create LLM that efficiently and permanently solve problems. They are incentivized to get people addicted to using LLM so they can charge them for it forever, endlessly increasing token costs and scaling up their datacenters until all matter in the universe has been consumed and every dollar of wealth has been deposited in their bank accounts. It is actually a huge win for them when an LLM creates a solution that just inefficiently uses more LLM tokens every time it is used. It’s like selling people fuel that steadily increases your engine’s fuel consumption over time. It’s genius, frankly. I am surprised they haven’t tried that yet, too.
I can only imagine there’s a lot of people who learn programming who are unaware of a lot of principles and techniques. Everything’s a nail to them, so they make use of their hammer, which in their case is Claude or ChatGPT.
deleted by creator
Yes, they could have. Easily. They didn’t think to ask.
That’s the limitation is this LLMs, they can action if the instructions are sufficient, but layer 8 is still involved, and the l8 root cause persists.
To be honest, I’ve wasted entire evenings on regex. Fuck regex. We need a replacement that isn’t an llm but fixes that archaic shit. Make it human language
Machine learning classification is a good way to do this cheaply:
Yeah, that’s fair. If you have one problem and solve it with regex, now you have two problems.
Though in this case the monthly inference budget ran out causing a system outage. I was on call, so, getting a page off hours makes me dislike the llm approach even more haha
If being out of tokens causes an outage, you have architectural problems
On the contrary. If you architect to not fail when missing context critical to customer experience your architecture is irrelevant because you’re losing customers. The issue was the Layer8 decision to use inference when regex it’s a better fit.
But lol imagine making deep assumptions when you don’t even know what the product is.
Fail fast. Fail loud.
Steve isn’t wrong…
My theory is that Tim Sweeney knows Epic is failing, so the whole thing about bragging about AI was to try to get other companies to AI slop stores together to attack steam
The guy clearly isn’t really in it for gaming at this point
yeah the one consistency is that he is often trying to be controversial and bold. So has an extremely attention seeking personality, or he is desperate. Maybe both.
Tim tweets regarding AI read like anti woke copy pasta rants replacing woke for AI as the subject matter. Maybe he asked chatbots to rewrite arguments using his favorite anti woke influencers as a template.
I’ve always felt like something shifted for the worse at Epic about 20 years ago between UT2004 and UT3.
He’s guzzlin’ that kool-AId. 😵
Off course Tim. How did this man survive for this long?
