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

Google CEO says more than a quarter of the company's new code is created by AI

businessinsider.com

Post reply on HN
Print view View original

161–170 of 1001 posts

Re: Google CEO says more than a quarter of the company's new code is created by AI

#161

Post by lysace »

kulahan wrote: post #145
lysace wrote: post #121

Github Copilot had an outage for me this morning. It was kind of shocking. I now believe this metric. :-) I'll be looking into ways of running a local LLM for this purpose (code assistance in VS Code). I'm already really impressed with various quite large models running on my 32 GB Mac Studio M2 Max via Ollama. It feels like having a locally running chatgpt.

I'm very happy to hear this; maybe it's finally time to buy a ton of ram for my PC! A local, private LLM would be great. I'd try talking to it about stuff I don't feel comfortable being on OpenAI's servers.

Getting lots of ram will let you run large models on the CPU, but it will be so slow.

The Apple Silicon Macs have this shared memory between CPU and GPU that let's the (relatively underpowered GPU, compared to a decent Nvidia GPU) run these models at decent speeds, compared with a CPU, when using llama.cpp.

This should all get dramatically better/faster/cheaper within a few years, I suspect. Capitalism will figure this one out.

Re: Google CEO says more than a quarter of the company's new code is created by AI

#162

Post by fzysingularity »

While I get the MBA-speak of lines-of-code that AI is now able to accomplish, it does make me think about their highly-curated internal codebase that makes them well placed to potentially get to 50% AI-generated code.

One common misconception is that all LLMs are the same. The models are trained the same, but trained on wildly different datasets. Google, and more specifically the Google codebase is arguably one of the most curated, and iterated on datasets in existence. This is a massive lever for Google to train their internal code-gen models, that realistically could easily replace any entry-level or junior developer.

- Code review is another dimension of the process of maintaining a codebase that we can expect huge improvements with LLMs. The highly-curated commentary on existing code / flawed diff / corrected diff that Google possesses give them an opportunity to build a whole set of new internal tools / infra that's extremely tailored to their own coding standard / culture.

Re: Google CEO says more than a quarter of the company's new code is created by AI

#163

Post by kulahan »

lysace wrote: post #161
kulahan wrote: post #145

Earlier quoted context omitted.

I'm very happy to hear this; maybe it's finally time to buy a ton of ram for my PC! A local, private LLM would be great. I'd try talking to it about stuff I don't feel comfortable being on OpenAI's servers.

Getting lots of ram will let you run large models on the CPU, but it will be so slow. The Apple Silicon Macs have this shared memory between CPU and GPU that let's the (relatively underpowered GPU, compared to a decent Nvidia GPU) run these models at decent speeds, compared with a CPU, when using llama.cpp. This should all get dramatically better/faster/cheaper within a few years, I suspect. Capitalism will figure th…

Interesting, so this is a Mac-specific solution? That's pretty cool.

I assume, then, that the primary goal would be to drop in the beefiest GPU possible when on windows/linux?

Re: Google CEO says more than a quarter of the company's new code is created by AI

#165

Post by lysace »

kulahan wrote: post #163
lysace wrote: post #161

Earlier quoted context omitted.

Getting lots of ram will let you run large models on the CPU, but it will be so slow. The Apple Silicon Macs have this shared memory between CPU and GPU that let's the (relatively underpowered GPU, compared to a decent Nvidia GPU) run these models at decent speeds, compared with a CPU, when using llama.cpp. This should all get dramatically better/faster/cheaper within a few years, I suspect. Capitalism will figure th…

Interesting, so this is a Mac-specific solution? That's pretty cool. I assume, then, that the primary goal would be to drop in the beefiest GPU possible when on windows/linux?

With Windows/Linux I think the issue is that NVidia is artificially limiting the amount of onboard RAM (they want to sell those devices for 10x more to openai, etc) and that AMD for whatever reason can't get their shit together.

I'm sure that there are other much more knowledgeable people here though, on this topic.

Re: Google CEO says more than a quarter of the company's new code is created by AI

#166

Post by yorwba »

seanmcdirmid wrote: post #129
ozim wrote: post #111

Earlier quoted context omitted.

That is what we pay managers -to figure out- for. They should find out which and how by knowing the team, familiarity with domain knowledge, understanding company dynamics, understanding customer, understanding market dynamics.

That's basically a non-answer. Measuring "productivity" is a well known hard problem, and managers haven't really figured it out...

Economists are generally fine with defining productivity as the ratio of aggregate outputs to aggregate inputs.

Measuring it is not the hard part.

The hard part is doing anything about it. If you can't attribute specific outputs to specific inputs, you don't know how to change inputs to maximize outputs. That's what managers need to do, but of course they're often just guessing.

Re: Google CEO says more than a quarter of the company's new code is created by AI

#167

Post by senko »

Twirrim wrote: post #138
ntulpule wrote: post #88

Hi, I lead the teams responsible for our internal developer tools, including AI features. We work very closely with Google DeepMind to adapt Gemini models for Google-scale coding and other Software Engineering usecases. Google has a unique, massive monorepo which poses a lot of fun challenges when it comes to deploying AI capabilities at scale. 1. We take a lot of care to make sure the AI recommendations are safe and…

> We work very closely with Google DeepMind to adapt Gemini models for Google-scale coding and other Software Engineering usecases. Considering how terrible and frequently broken the code that the public facing Gemini produces, I'll have to be honest that that kind of scares me. Gemini frequently fails at some fairly basic stuff, even in popular languages where it would have had a lot of source material to work from;…

We collectively deride leetcoding interviews yet ask AI to flawlessly solve leetcode questions.

I bet I'd make more errors on my first try at it.

Re: Google CEO says more than a quarter of the company's new code is created by AI

#168

Post by Smaug123 »

airstrike wrote: post #130
skrebbel wrote: post #127

To my experience, AIs can generate perfectly good code relatively easy things, the kind you might as well copy&paste from stackoverflow, and they'll very confidently generate subtly wrong code for anything that's non-trivial for an experienced programmer to write. How do people deal with this? I simply don't understand the value proposition. Does Google now have 25% subtly wrong code? Or do they have 25% trivial code…

By definition, "trivial" code should make up a significant portion of any code base, so perhaps the 25% is precisely the bit that is trivial and easily automated.

I don't think the word "definition" means what you think it means!

Re: Google CEO says more than a quarter of the company's new code is created by AI

#169

Post by chongli »

toasteros wrote: post #144
skrebbel wrote: post #127

To my experience, AIs can generate perfectly good code relatively easy things, the kind you might as well copy&paste from stackoverflow, and they'll very confidently generate subtly wrong code for anything that's non-trivial for an experienced programmer to write. How do people deal with this? I simply don't understand the value proposition. Does Google now have 25% subtly wrong code? Or do they have 25% trivial code…

> the kind you might as well copy&paste from stackoverflow This bothers me. I completely understand the conversational aspect - "what approach might work for this?", "how could we reduce the crud in this function?" - it worked a lot for me last year when I tried learning C. But the vast majority of AI use that I see is...not that. It's just glorified, very expensive search. We are willing to burn far, far more fuel t…

we've decided we can't be bothered with traditional search

Traditional search (at least on the web) is dying. The entire edifice is drowning under a rapidly rising tide of spam and scam sites. No one, including Google, knows what to do about it so we're punting on the whole project and hoping AI will swoop in like deus ex machina and save the day.

Post reply on HN