Posted by therepanic 6 hours ago
> AI is something that’s happening mostly due to Moore’s law and general progress in computing, not something that they are doing.
But if these companies control the vast majority of compute power, which seems like the plan they are already executing, won't they capture most of the value from the progress of AI?
> And two, this strawman jump from, oh hey, it’s a fancy autocomplete, smart compiler, better search engine, to it’s gonna like own the whole light cone bro like if you aren’t in SF and at the right parties there’s gonna be like a flash of light in the sky one day and you’re not even gonna know what happened but everything just Changed.
Haha, OP has a way with words.
In a way, both these emotional extremes (FOMO & the singularity) are just tools being used to continue driving the massive CapEx behind LLM improvement. Hate to love it? Love to hate it?
It's bullshit in the sense that they don't know for sure, but the author doesn't either. Why might or might not it be true?
May not be true because it's a blind spot to assume that purely by being a player in the AI game (with no real attention paid to quality of result), you have increased odds of winning the game. That's true in the abstract, but practically, it requires a competent player to become true in reality.
The blog has a tagline, "the singularity is nearer". I think belief in a "singularity" almost implies these things to some degree.
Wait, does this mean I'm better at something than geohot? All that time spent learning regexps wasn't a waste!
This is what he wrote before.
> I’m calling it now, the adoption of AI agents into software development will be one of the most costly mistakes in the field’s history. Agents cannot program, and it’s taking longer and longer to realize that they can’t.
Now he's writng
> I love the progress. I’m so excited for the new LLMs, self driving cars, video generation models, and coding agents.
SMH now he writes about the hype. My brother in absolute Deity, *you* should have believed the hype.
He does say in this post:
> I’m getting better at using them and get some boost from the models. It is a new skill, and it’s not like I haven’t constantly been trying them. You have to be really careful, they can increase cognitive fatigue, and all the vibe coded stuff is still slop (where’s all this new magical software that the productivity improvements should imply?).
"Where’s all this new magical software that the productivity improvements should imply?"
This is a recurring gotcha in the anti-AI marketplace of denialism. It's a bit like saying "I saw a fat guy, so why do people keep telling me that GLP-1s change everything?"
It takes time. Like already I would say just about every programmer has replaced a number of tools with random shit they spit out from LLMs. It percolates out from there as some things become products, etc.
And to anyone actually paying attention, and not just feeding their delusions, the impact is utterly enormous. Incontestable. The "Where's the software? / Where's the change?" people are absolutely going to find themselves in the dustbin of history.
It's also fascinating how often people do the stochastic parrot horseshit.
The other night I had to do a large scale compression test with libjxl, which notably is software that has seen an enormous amount of optimization interest and you would assume has little extra to be eked out. I've traced through this software before and the compression path is insanely complex. It's the sort of software that is headache inducing. Anyways, curious what the state of the platform was I grabbed the latest source and asked Fable to look for low-hanging fruit in the lossy and lossless compression paths. It suggested a few, created a test harness to A:B bitwise compare with the original, and implemented its optimizations. It achieved a 14% performance increase in a single pass, using just the remaining quota I had on a subscription as my week drew to a close. And all it did was some high level logical optimizations, some more efficient memory allocations, and so on. All of its code was completely in the style of the project, was no more significant than necessary, and so on. Anyone that isn't utterly blown away by that -- who gets the hype -- is lying to themselves.
It’s why con artists, scammers always flood every hype cycle. Greed ruins everything.