prettyblocks

I'm more surprised that people think using AI automatically makes image generation not art. AI tooling can be quite complex, and it's much harder than most people think to produce consistent, compelling results, especially with local setups on modest hardware.

It's still there.

Isn't that typical of Nathan's work? To mess with the viewer.

NSO was sanctioned, so slightly higher than zero imo.

In this book, they explain that it's because this industry is de-regulated:

Pegasus: How a Spy in Your Pocket Threatens the End of Privacy, Dignity, and Democracy by Lauret Richard

I did the exact same project about a month ago with Opus4.8. No Asta required.

It's hard to get a $90k bounty from Google, even with Claude in hand, so we shouldn't be dismissive of that, but it is pretty annoying to read write-ups written entirely in LLM voice.

I just had claude code set it up for me and create some startup scripts. The longest part was the model download.

I've been playing with this on a 3090 and it FLIES. Does a pretty good job too on the tasks I've thrown at it (php code base security audits).

The first time I heard about Burning Man was at Defcon back in the early 00's. Some folks were telling me how they're going after the con and I asked, what's Burning Man? They told me how it was a sick festival in the desert where people race cars, shoot guns and set stuff on fire. By the time I made it in '09, it was tens of thousands of people, crawling with feds, and a strict 5mph limit was imposed on all vehicles. I had a great time though.

The abliterated qwen3.8 has been really good too

I remember when they promised to slow down and released a new model less than a week later.

I generate them using LLMs, but optimize them by manually removing chunks or rearranging the order of the instructions. It works well.

They're pushing their customers to their own competition by doing this.

the worry is that this will be interpreted as tampering and cary consequences.

> when you can just start from zero, look at what's known to be happening right now, and draw a very straight line to catastrophe.

The problem is that it's not a straight line at all. "We are failing at alignment" is a valid concern, but there's not a straight line to "we will all be extinct within x year", and it's always a wild hyperbolic leap.

Do you assume that every post criticizing anything is part of a sockpuppet disinformation campaign as well?

I guess that's one way to guarantee development slows down to a crawl, but in no way would it increase trust.

It depends what you mean by meaningful, but yes, qwen3.8 27b is pretty mind blowing to me. It can easily solve portswigger labs for example at q4_k_m.

strange assertion as this has been an ongoing problem from day one.

Do you use a 3rd party provider or use deepseek directly?

As someone who gets a ton of these inbounds every day, I report every single one of them as spam.

I have a theory that when someone is ready to resign from a frontier lab they must get offered a substantial bonus to post the most unhinged doomer take they can think of.

I use skills to automate non-development workflows, like for malware analysis, etc.

I find them very useful.

Yet the products they release are purposely dumbed down in the name of alignment. I'm in the CVP and Fable downgrades most of my work to Opus, it's incredibly frustrating.

It's probably a good model for folks doing basic software stuff, or humanities related tasks, but I work in cybersecurity on the defense/detections side and I haven't been able to use it for anything even with being in the CVP. It downgrades to Opus every time.

I haven't been in about 10 years, but yes it's incredibly fun. What you see on social medial is not a representation of the majority of attendees, but it is expensive to attend, especially if you need to travel for it.

You hear that everyone? Just stop after $30 mil and you will be happier.

No opinion on the salary, but just because it's a once per year event doesn't mean the work isn't year round.