Earlier discussion: A deep dive into self-improving AI and the Darwin-Gödel Machine https://news.ycombinator.com/item?id=44174856

There’s definitely been plenty of denuvo games cracked, but I’d say most games that haven’t been cracked have denuvo. I think it also depends on the version of denuvo. Newer versions seem pretty well protected

Well done terraform modules are super powerful. I’m forced to use cdk at work and holy shit is that so much worse. really not easy to pass modules or constructs around. Especially if you are switching between languages

Haven’t heard this framed this way before. I feel like AI development is gonna make this stuff way more common. Would love to see AI built to tackle this stuff, but seems unlikely for now. In theory an AI agent would be a good dummy user as they forget everything about the app with each iteration, but might be hard to get the scaffolding right

Keep up the great work! I can’t wait for their Orion browser to come to Linux. Would also love to see it properly open sourced

There is politics in the original trilogy. It’s just more thematic than literal political manoeuvring. The story of a rag tag group of rebels fighting against a massive entrenched empire is a political one by nature. And Lucas has directly stated his inspiration for that dynamic was the Vietnam war. And spoiler alert America isn’t the rebels. I think Andor does a great job of fleshing out the more day to day feelings of being in an era of rebellion against an empire bent on domination

The Simpsons had it right when they parodied Episode 2 as a boring story about redistricting.

What’s the compute requirements, or like throughput min and max?

Yeah I’d love to see this replicated across various system prompts as well. They make a good point at the end that the system prompts encouraged high morality and high agency. I’m wondering if you just did one or the other, or neither if they’d exhibit the same behaviour.

It does feel like an ASI with access to all the data and ideas around climate change would basically inevitably come to this conclusion. Perhaps it could change the exact definition of what’s in our means with more efficient resource production, recycling, and distribution, but at the end of the day it’s a fairly simple inputs vs outputs usually.

I see you put how to calculate the compression ratio, but do you have some examples of worst case, average, and best case compression ratios?

Edit: ok I see the photos in the repo. Putting them in the README would be helpful

Author here. The repo is a complete mess but I do have some some code in there to generate graphs and whatnot if you're willing to dig through the code. I will make this much more concrete with lots of proper testing. Its very much still a messy work in progress.

I applaud you for uploading even if it’s a bit disorganised still, and I do the same. It’s better to have something than nothing. Sometimes I see people talking about something but they don’t want to upload their code yet because they have to clean it up first. And then either they don’t get around to ever cleaning it up the way they wanted, or by they time they do it will have fallen off everyone’s radar and be forgotten. At least with a messy repo it’s possible to poke around, and not the least to star it and maybe check back later to see if they cleaned it up.