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

Advent of Code 2023's new AI/LLM Policy

adventofcode.com

Post reply on HN
Print view View original

111–120 of 133 posts

Re: Advent of Code 2023's new AI/LLM Policy

#111

Post by reidjs »

danjoredd wrote: post #86
postalrat wrote: post #4

"(If you want to use AI to help you solve puzzles, I can't really stop you, but I feel like it's harder to get better at programming if you ask an AI to do the programming for you.)" I'd say that whatever "better at programming" is quickly evolving. AI is a tool like high level languages, IDEs, autocomplete or all the different types of code analyzers. How is ignoring a powerful tool like AI going to get you better a…

For the same reason giving a 5 year old a calculator to "learn" addition doesn't teach them math. You still need to know why shit works in order to use it in a fully educated way. Otherwise you are just "putting magic words into the text editor" and your code will be held together with bailing wire and chewing gum. For day-to-day use, as long as you know what the code is doing when you run it, its fine. But to get to…

I’m not a 5 year old, I’m an old guy who has a job to finish. That said, it depends what you want to get out of AoC. for me, the prize is to learn something by solving the puzzles, so I probably wouldn’t plug the problem statement into the llm directly

Re: Advent of Code 2023's new AI/LLM Policy

#112

Post by simonw »

withinboredom wrote: post #101
polygamous_bat wrote: post #92

Earlier quoted context omitted.

I am curious, why is this an illusion? I have experienced the same (or at least similar) feeling of getting better after getting my hands on a search engine like google. I never stopped using a search engine since then to check if I am suffering from this “illusion” you’re talking about. Should we?

I seriously suggest getting a book and typing out a listing totally from scratch. It's interesting to actually type out another person's style of solving a problem because not only is it alien feeling to "solve it that way", it's also interesting to physically solve it that way. I did this recently and was surprised to see myself solving the problem "that way" at work. It was much more elegant and more readable than…

What was the book?

Re: Advent of Code 2023's new AI/LLM Policy

#113

Post by datadrivenangel »

HKH2 wrote: post #69
lgeorget wrote: post #48

Earlier quoted context omitted.

The incentives to cheat in a professional competition are more obvious since the money you earn is related to your performance. In the case of AoC, the only objective incentive I can think about is impressing a recruiter. Maybe it's enough if you're motivated enough.

Why do people cheat in video games?

Money. Being really good at a game can be a source of income for E-sports or streaming.

Re: Advent of Code 2023's new AI/LLM Policy

#114

Post by withinboredom »

simonw wrote: post #112
withinboredom wrote: post #101

Earlier quoted context omitted.

I seriously suggest getting a book and typing out a listing totally from scratch. It's interesting to actually type out another person's style of solving a problem because not only is it alien feeling to "solve it that way", it's also interesting to physically solve it that way. I did this recently and was surprised to see myself solving the problem "that way" at work. It was much more elegant and more readable than…

What was the book?

https://craftinginterpreters.com/

Highly recommend it. Very much.

Re: Advent of Code 2023's new AI/LLM Policy

#115

Post by forgotpwd16 »

lsy wrote: post #22

Maybe I'm a dinosaur but I'm utterly mystified by folks who are interested in making the leaderboard or even solving AoC puzzles with AI assistance. To me it feels like going to a 5k charity run, taking a motorcycle to the finish line, and parading your gold medal around. Like, who are you impressing? And are you even proud of yourself? I understand the appeal of using the best and fastest means when there are real-w…

It's just like cheating in online games. Heck people will cheat even on games with no ranking at stake.

Re: Advent of Code 2023's new AI/LLM Policy

#116

Post by michaelteter »

lsy wrote: post #22

Maybe I'm a dinosaur but I'm utterly mystified by folks who are interested in making the leaderboard or even solving AoC puzzles with AI assistance. To me it feels like going to a 5k charity run, taking a motorcycle to the finish line, and parading your gold medal around. Like, who are you impressing? And are you even proud of yourself? I understand the appeal of using the best and fastest means when there are real-w…

> mystified by folks who are interested in making the leaderboard or even solving AoC puzzles with AI assistance

I think it defeats the purpose of the AoC to use AI, but I can imagine for some people it's a different game to see how they can use a screw (gaming term) to circumvent the system.

On the other hand, there may be some value to learning how to apply AI tools as a solver. If test cases are available for a given problem space, and AI can be applied to build code to meet those "requirements", then it is probably a skill we should be learning.

While I don't believe people are going to play fair here, it would be nice if AoC submissions had an [AI-assist] checkbox. Then there could be two leaderboards, and the AI folks could still have some good competition with each other.

Ultimately, the AoC leaderboards are unfair as they favor people in timezones that fit the release of each "day". So being on the leaderboard isn't really something that many people can do, even if they are brilliant problem solvers.

Re: Advent of Code 2023's new AI/LLM Policy

#117

Post by bunderbunder »

umvi wrote: post #105
jasonjmcghee wrote: post #13

If you’re competing in a whittling competition, I’m guessing CNC machines are discouraged. Seems sensible.

And furthermore (in AoC's opinion), being reliant on CNC machines make you worse at whittling, which I guess makes sense.

I don't think it's necessarily that it makes you worse; it's that it defeats the purpose of the exercise.

Like, I'm sure there are some professional CNC operators who are also very, very good at doing things by hand. But they didn't get good at doing things by hand by using a CNC machine to do everything that can be done on a CNC machine. And, based on what I observed growing up in a down-and-out manufacturing town, the people who decided that CNC is the way of the future and went all in on just being very, very good CNC operators are the ones who got hit hardest by the growth of industrial automation in the 1990s and 2000s.

I have similar concerns about Copilot. In the short term, yes, it lets you get relatively straightforward tasks done more quickly. But I suspect that people who make heavy use of Copilot risk plateauing at a lower skill level and ultimately being the people whose jobs are most vulnerable to the growth of technologies like Copilot.

Re: Advent of Code 2023's new AI/LLM Policy

#118

Post by michaelteter »

dghlsakjg wrote: post #78
HPsquared wrote: post #37

Earlier quoted context omitted.

People can tell if you use a motorbike though. More like, performance enhancing drugs... or designing an anthropomorphic running robot that completes the race for you. Designing such a robot would be an achievement in its own right, but not really in the spirit of the competition.

If you designed and built the robot, its one thing (still not right, but interesting). If you just rented some time on someone else's robot, its just plain cheating. I would love to see teams compete, in a separate venue/ranking, based on their own model's ability to write code with minimal prompting. It would be kinda cool to see what can be done in different 'classes' of model. How far can you get with a local mode…

> I would love to see teams compete, in a separate venue/ranking, based on their own model's ability to write code with minimal prompting.

Yes, an AIdvent of Code would be interesting...

Re: Advent of Code 2023's new AI/LLM Policy

#119

Post by lionkor »

I'm perfectly happy to let new programmers use AI to learn, that way I'm very cozy in any job I'm in.

Nothing to worry about if new programmers are learning from AI that writes, on average, code much worse than anyone I know.

There was always the opion of not doing your work yourself, and instead paying a few bucks on some site to have someone else do it. Same crowd, it just got cheaper.

Post reply on HN