It's happened once, could happen any time.

Not good for business!

https://red.anthropic.com/2026/n-days/

So that is a poor bandaid to use now. Maybe instead validate things before, and have more of a cathedral and human reputation system.

This is not complicated. Even in the US, every other industry is regulated to your benefit, you're just used to it and haven't realised. Digital technology obviously needs to be too. And yes, you have to do it properly.

This old post goes into lots of detail about what they do to red team and why: https://www.aisi.gov.uk/blog/early-lessons-from-evaluating-f...

NIST's similar unit in the US is now called CAISI https://www.nist.gov/caisi - interesting that the most recent post is an evaluation of DeepSeek capabilities, which sound more like watching China. But presumably this executive order alters the emphasis?

That could easily be true and Apple "will use even more tokens and spend even more money".

Avoid lock in to stack from one provider (things like a harness that only works with models from one provider and so on).

Use local models (a couple of them do work a bit now, if you have 20Gb video RAM), which saves money and is more private, and works offline.

Can improve the harness, fix bugs in it, make it compatible with different systems and techniques.

This game happens every time in new cycles of developer technology. The good bet historically has always been to use open source - there's a reason most developer tooling just pre-AI revolution was open source (even things like Java and .NET which used to be proprietary).

https://rsf.org/en/index

Until I got to "One platform, three modes." and my brain just pattern matched "AI slop" and the entire post dissolved into meaningless for me.

I don't know if I can stop my mind reaching this conclusion. I'm sure someone at GitLab made some effort to carefully edit the post... But that it wasn't entirely rooted in a human who'd worked out how this stuff goes, but clearly had lots of AI writing it out... Just made my instinct go "this isn't worth paying attention to after all".

They've been decent at full rewrite for 2 years. I don't think they were good at search/replace until a year ago, but I'm not so sure.

It's true that the models 2 years ago would sometimes make errors in whole rewrite - e.g removing comments was fairly common. But I've never seen one randomly remove one character or anything like that. These days they're really good.

Main reason agentic harnesses use search/replace is speed and cost, surely! Whole file output is expensive for small changes.

I'm finding Google's Gemma 4 even better though - seems to hold up the agentic loop better than Qwen.

All will load into 20Gb of VRAM. None are amazing, but they do just about work.

Comments URL: https://news.ycombinator.com/item?id=47922996

Points: 3

# Comments: 1

https://github.com/twitter/the-algorithm/blob/7f90d0ca342b92...

When this started it really put me off X - I'd have tolerated, and almost liked the idea, of a freedom of speeech place. But a place that boosts its owners posts... Nope.

I'm out - it's such a big personal diss of me, I'm not interested any more.