Live data from Hacker News

Codex, Opus, Gemini try to build Counter Strike

instantdb.com

71–80 of 144 posts

Re: Codex, Opus, Gemini try to build Counter Strike

#71

Earlier quoted context omitted.

Personally i find it absurd that code can be copyrighted at all.

Copyright is so-so. At the end of the day you can say that the complete work (not just snippets) is something copyrightable. But the most bananas thing for me is that one can patent the concept of one click purchasing. That's insane on many levels.

Why bananas? That is the biggest invention after edisons bulb.

Re: Codex, Opus, Gemini try to build Counter Strike

#72
post #55

Earlier quoted context omitted.

> Nobody is going to sue you for using their implementation of a skybox algorithm from 1999, give us break. For personal use maybe not, but that's not the point, the point is it's spitting out licensed code and not even letting you know. Now if you're a business who hire exclusively "vibe" coders with zero experience with enterprise software, now you're on the hook and most likely will be sued.

Do you have any evidence that it is spitting out licensed code? Did you locate an original that it was copied from?

This particular case appears to me to be a straight derivative at best but I'm by no means an expert on copyright laws.

That's not to say there hasn't already been more direct cases with set examples [1], from an author directly who would have a better right to claim than I [2], it's not even a stretch to see how it can happen.

[1] https://arxiv.org/html/2408.02487v3

[2] https://x.com/DocSparse/status/1581461734665367554

Re: Codex, Opus, Gemini try to build Counter Strike

#73
This is the kind of thing that's so impressive that if you're not an (experienced) SWE you think "man LLMs are the future, and I am making some major decisions based on this". But you look at the code, and it's essentially gluing three.js and some DB stuff together. There's no lobby, no real interaction logic, no physics apart from what you get from three.js, chatting, commands, map editing, game modes.

In other words, this is slop. We know these new models can generate slop images, text, videos, and code. Sometimes slop can be useful; maybe you can shape it into something useful, maybe you can slop a slopper. But we're learning it's not economical--this is some of the costliest slop we've ever made.

Re: Codex, Opus, Gemini try to build Counter Strike

#74
post #63

This makes me depressed. LLMs will take the most enjoyable part of my job and I will be stuck reviewing or fixing bugs in their "it-compiles" codebase.

On the contrary, coding for me has become more fun than ever since Opus 4.5. I'm working more and genuinely enjoying it a lot more, haven't had this much fum building software in years. (I work at Anthropic but have also tried Gemini, it's also fun)

Yes, lived experiences differ from person to person.

Re: Codex, Opus, Gemini try to build Counter Strike

#75
post #8

I tried to find some code that wasn't minified to assess the quality of this, and I found some shader code for the sky in the gemini version. The whole shader looks like it was regurgitated verbatim. This wouldn't hold up to licensing scrutiny. Here's a snippet from it: // wavelength of used primaries, according to preetham const vec3 lambda = vec3( 680E-9, 550E-9, 450E-9 ); // this pre-calcuation replaces older Tota…

I also noticed that AI agents commit many copyright infringements with the work of Mr Dijkstra.

Messrs Newton and Raphson would like to join this class-action.

Re: Codex, Opus, Gemini try to build Counter Strike

#76
post #40
post #27

Earlier quoted context omitted.

I always find it amazing that people are wiling to use AI beacuse of stuff like this, its been illegally trained on code that it does not have the license to use, and constantly willy nilly regurgitates entire snippets completely violating the terms of use Edit: https://github.com/vorg/pragmatic-pbr/blob/master/local_modu... https://github.com/vorg/pragmatic-pbr/blob/master/local_modu... This looks like where the sou…

As discussed in this thread before you posted this comment, this code wasn't generated from an LLM at all, but simply included in a dependency: https://news.ycombinator.com/item?id=46092904 Unlike your results which aren't exact match, or likely even a close enough match to be copyright infringment if the LLM was inspired by them (consider that copyright doesn't protect functional elements), an exact match of the cod…

There are several cases where copyright law is not only about exact copy but also derivatives. So finding an exact match is not necessary.. Not sure it matters in this case eitherway.

Re: Codex, Opus, Gemini try to build Counter Strike

#77
post #73

This is the kind of thing that's so impressive that if you're not an (experienced) SWE you think "man LLMs are the future, and I am making some major decisions based on this". But you look at the code, and it's essentially gluing three.js and some DB stuff together. There's no lobby, no real interaction logic, no physics apart from what you get from three.js, chatting, commands, map editing, game modes. In other word…

This is the job a junior developer may deliver in their first weeks at a new job, so this is the way it should be treated as: good intentions, not really good quality.

AI coding needs someone behind to steer it to do better, and in some cases, it does. But still hasn't left the junior phase, and while that doesn't happen, there's still the need for a good developer to deliver good results.

Re: Codex, Opus, Gemini try to build Counter Strike

#78
post #44

This is very impressive. That said, 1st person shooters seem like the less interesting type of game to create with an LLM nowadays. I'd much rather see a large world mystery game, for example. Think something like "All Her Fault", where you're the mom and you show up to pick up your kid and the game starts there -- and you need to find your kid. I would fine a game like that something that we probably couldn't do wel…

The most impressive thing about this wave of AI is the speed of the goal posts.

how exactly are the goal posts moving?

the code and output is literal slop

it's not known how much editing and debugging was done by the team either

you could have done this in 2022 with not that much debugging as well

Re: Codex, Opus, Gemini try to build Counter Strike

#79
post #56

Earlier quoted context omitted.

Anthropic shared some approximate numbers on their Smart Contracts post. A 5h research assignment currently has an API cost of ~3.500USD.

What does 3.500USD mean? Are we bringing back the half-penny?

He’s European and uses a “.” instead of a “,”
Post reply on HN