Took me a second to realise the title’s refering to “AI slop”, my inital reaction was something like “I know, it’s always been slop”.

They didn’t make a video about it because they thought it was a problem for creators, not a problem for consumers.

Which is true. Influencers are great at making their thing your thing, because that’s kind of their job, and we’ve seen it many times before. Just look at all the outrage about the YouTube algorithm and such, it doesn’t matter to anyone except influencers but somehow it’s made to be everybody’s business.

This feels very similar. Scummy business practice, good on them for suing, but to the rest of us it should only be a curiosity.

Isn’t this kinda what the controversy around the ElastiSearch licensing change was about? I think people have had similar frustrations with HashiCorp software, but I don’t know the details.

I think you’re spot on with LLMs being mostly trained on these kinds of tasks. Can’t say I’m an expert in how to build a training set, but I imagine it’s quite easy to do with these kinds of problems because it’s easy to classify a solution as correct or incorrect. This is in contrast to larger problems which are less guided by algorithmic efficiency and more by sound design/architecture.

Still, I think it’s quite impressive. You don’t have to go very far back in time to have top of the line LLMs unable to solve these kinds of problems.

Also there is no big consequence if they don’t and it’s probably possible to bruteforce (which is how many programming tasks have been solved).

Usually with AoC part 1 is brute-forceable, but part 2 is not. Very often part 1 is to find the 100th number, and part 2 is to find the 1 000 000 000 000th number or something. Last year, out of curiosity, I had a brute-force solution for one problem that successfully completed on ~90% of the input. Solution was multi-threaded and running on a 16 core CPU for about 20 days before I gave up. But the LLMs this year (not sure if this was a problem last year) are in the top list of fastest users to solve the problems.

I hardly see it changed to be honest. I work in the field too and I can imagine LLMs being good at producing decent boilerplate straight out of documentation, but nothing more complex than that.

I think one of the top lists on advent of code this year is a cheater that fully automated the solutions using LLMs. Not sure which LLM though, I use LLMs quite a bit and ChatGPT 4o frequently tells me nonsense like “perhaps subtracting by zero is affecting your results” (issues I thought were already gone in GPT 4, but I guess not, Sonnet 3.5 does a bit better in this regard).

Why are they doing this? Seems detrimental to Russian businesses and also sours relations with their only “friend” in the whole world?

Also not an issue if you’re in the rich part of the world, or just one that has a lot of water. Fortunately I don’t think water is gonna be what makes Russia invade, don’t know what their supply looks like but I can’t imagine it’s not enough.

oh, I thought he didn’t care?

Didn’t the refactored netscape eventually evolve into Firefox though? Not disputing the poster child status or the fact that it’s a terrible business decision, but the project did not really go stale I think?

What’s IVF in this context?

What kind of drones are these? Sounds like they’re consumed when used, are they basically missiles/rockets with better control?