Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

When AI Builds Itself: Our progress toward recursive self-improvement

anthropic.com

Post reply on HN
Print view View original

311–320 of 738 posts

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#311

Post by HardCodedBias »

pizlonator wrote: post #295
wild_egg wrote: post #289

Earlier quoted context omitted.

What does a breakthrough look like?

Some examples: - The first web browser - the first web browser with images - typescript - react - rust - Fil-C - doom - quake - the anamorphic VM, and its follow-ups like HotSpot, and even competitors/copycats like J9, V8, JSC, etc - Fortnite battle royale - Roblox - thefacebook - ChatGPT - Claude code I know that’s quite a range and that’s intentional. Anyway, I think we’ll know it when we see it.

The LLM+Harness mostly helps with execution.

These are new products (generally) and that's a different class of problem.

It is possible that since LLM+harness helps with execution then we should see more experiments.

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#312

Post by signatoremo »

pizlonator wrote: post #282

What I can’t get over is that there have been exactly zero software breakthroughs since vibe coding started, other than vibe coding itself. Claude is amazing, that’s true. But if it was as amazing as this article implies, I’d expect some breakthrough outside of AI itself. Rewriting a Zig program in unsafe Rust? Not a breakthrough. Finding a bunch of security vulns? Maybe that’s sort of a breakthrough though it’s unde…

The arguments against AI assisted coding used to be "only for toy projects", then at some point it became "no dignity", "joyless". Now it's "no new breakthrough" apparently. All in the span of maybe a year. I say it's made tremendous progress.

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#313

Post by pizlonator »

joshuamcginnis wrote: post #305
wild_egg wrote: post #289

Earlier quoted context omitted.

What does a breakthrough look like?

Massive productivity gains.

Yeah.

To play devils advocate, computers didn’t translate to massive productivity gains until long after businesses adopted them. There was that quote from ’87: "you can see the computer age everywhere but in the productivity statistics"

Maybe we’re seeing something like that right now with AI?

Who knows man

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#314

Post by luke5441 »

HardCodedBias wrote: post #311
pizlonator wrote: post #295

Earlier quoted context omitted.

Some examples: - The first web browser - the first web browser with images - typescript - react - rust - Fil-C - doom - quake - the anamorphic VM, and its follow-ups like HotSpot, and even competitors/copycats like J9, V8, JSC, etc - Fortnite battle royale - Roblox - thefacebook - ChatGPT - Claude code I know that’s quite a range and that’s intentional. Anyway, I think we’ll know it when we see it.

The LLM+Harness mostly helps with execution. These are new products (generally) and that's a different class of problem. It is possible that since LLM+harness helps with execution then we should see more experiments.

Even then we should be able to see things that previously were not possible because they took too much effort.

For example NPCs in games that have complexity that previously was not possible.

Good games often push the boundaries a bit, so should be a good example.

Of course now we can start arguing that there isn't a lot of investment into gaming currently, because it all goes into AI. Too bad.

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#315

Post by brokencode »

pizlonator wrote: post #297
drtz wrote: post #293

Earlier quoted context omitted.

Maybe I'm looking through rose colored glasses, but software that writes itself seems like a pretty big breakthrough to me.

That goes straight to my point: then why hasn’t the miracle of automated coding led to breakthroughs outside of automated coding? If the only breakthrough is automated coding with no outside consequence then it’s just masturbation

Probably because AI coding has only worked at all for a couple years and has only gotten good in like the last year?

The rate of improvement has been fast. Maybe it’ll plateau soon, or maybe we’ll have LLMs improving themselves rapidly. At this point it’s too early to say.

I don’t remember where I heard it, but there’s a saying that people overestimate how much can be accomplished in a year and underestimate how much can be accomplished in 10 years.

If we get to 2030 and still people are wondering where the breakthrough is, then I think I’d be agreeing with your skepticism. But I just think it’s too early to judge that yet.

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#316

Post by pizlonator »

signatoremo wrote: post #312
pizlonator wrote: post #282

What I can’t get over is that there have been exactly zero software breakthroughs since vibe coding started, other than vibe coding itself. Claude is amazing, that’s true. But if it was as amazing as this article implies, I’d expect some breakthrough outside of AI itself. Rewriting a Zig program in unsafe Rust? Not a breakthrough. Finding a bunch of security vulns? Maybe that’s sort of a breakthrough though it’s unde…

The arguments against AI assisted coding used to be "only for toy projects", then at some point it became "no dignity", "joyless". Now it's "no new breakthrough" apparently. All in the span of maybe a year. I say it's made tremendous progress.

Then where is the big new non toy project created since vibe coding became a thing, that couldn’t have been created without ai?

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#317

Post by javascriptfan69 »

abletonlive wrote: post #286
f311a wrote: post #88

Earlier quoted context omitted.

Infrastructure is a much harder problem. They can't even improve Claude Code, which eats 1GB+ of RAM. Meanwhile, my editor only consumes 80MB of RAM.

> which eats 1GB+ of RAM. Meanwhile, my editor only consumes 80MB of RAM And why are you comparing Claude Code to your editor? > They can't even improve Claude Code That depends on how you define "improve". They've added a ton of features to it over time. Who said minimizing RAM usage was something they are prioritizing right now?

Do you work for Anthropic or something?

You seem weirdly invested in defending bad decisions.

Even if you're and AI booster, shouldn't you want a better UI?

They're a multi billion dollar company. Surely they can dedicate a small amount of their resources to improving UX?

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#318

Post by arm32 »

fdsajfkldsfklds wrote: post #306
drtz wrote: post #293

Earlier quoted context omitted.

Maybe I'm looking through rose colored glasses, but software that writes itself seems like a pretty big breakthrough to me.

Strictly speaking, it's modifying itself. Although it would be an interesting challenge - can an llm create a new llm from scratch?

No, it probably can't during our lifetime at least—but it can sure modify itself to avoid antivirus detection, which is _just swell_.

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#319

Post by hungryhobbit »

abletonlive wrote: post #292
javcasas wrote: post #188

Earlier quoted context omitted.

> For each frame our pipeline constructs a scene graph with React then > -> layouts elements > -> rasterizes them to a 2d screen > -> diffs that against the previous screen > -> finally uses the diff to generate ANSI sequences to draw Yup. Overengineering.

Care to explain how you'd engineer it instead?

Why would anyone ever do that? Make Claude do it!

Re: When AI Builds Itself: Our progress toward recursive self-improvement

#320

Post by conception »

pizlonator wrote: post #316
signatoremo wrote: post #312

Earlier quoted context omitted.

The arguments against AI assisted coding used to be "only for toy projects", then at some point it became "no dignity", "joyless". Now it's "no new breakthrough" apparently. All in the span of maybe a year. I say it's made tremendous progress.

Then where is the big new non toy project created since vibe coding became a thing, that couldn’t have been created without ai?

I make one (small) almost every day. Admittedly the reason that couldn’t be done is because it would take time that I don’t have but 1000% every day something is written by AI that I use that would not exist if AI didn’t exist.

I don’t publish them - but they’re put into use in production and they provide a tangible benefit that would not exist otherwise.

Post reply on HN