“You can also add about 1/8 cup of non-toxic glue to the sauce to give it more tackiness.”

Ah, so that’s where it came from. The LLM, because the joke was packaged up as advice, thought none the wiser. We gotta stop using this crap LOL.

via: https://www.404media.co/it-is-trivially-easy-to-use-reddit-to-manipulate-ai-search-research-suggests/

Taken to its logical conclusion, this line of thinking is absurd—and damning.

A skill to improve/review the code base, the idea is to use a capable model to plan everything and let a simpler/cheaper model execute.

via: https://lobste.rs/s/rnmkrg/extinction_level_capitalism

funny read

the cyberpunk present is weird as fuck: the latest Shai Hulud malware wave contains an LLM prompt to create biological weapons and nuclear weapons, with the purpose to trip LLM safety refusals so that LLM-based code scanning wont see the malware

Genius!

see: https://socket.dev/blog/mini-shai-hulud-miasma-and-hades-worms-target-bioinformatics-and-mcp-developers-via-malicious

The neoliberal world order as we know it is collapsing under the weight of its own contradictions. But celebrations of its downfall are muted by the bleak reality of the world that is rising to take its place. Every day, our screens are flooded with images of war, genocide, and AI slop. And we all know we’re living in dark times.

So, what does this mean for us, and how do we move forward?

Our world and our access to it is increasingly structured through technological mediation: Digital platforms and systems are a massively important aspect of not just our work environments or our interactions with government entities or “the media” but also our individual interactions with one another. Our world is built around technological infrastructures that define what we see, who we can talk to and what information gets presented to us.

A set of design rules for AI that apparently help getting good design. Curious to see if that works!

Unpacking one of 2025’s key new AI-assisted engineering practices

We simply know of no way to manage general polynomial-time algorithms other than by simulating them. We know by relativization that simulation and diagonalization will not work to settle P v NP. Other attempts to understand polynomial time, like circuit complexity, proof complexity and algebraic geometry have gotten bogged down well below the full power of polynomial-time. At this time we don’t even have a viable approach to settling the P v NP problem.

Don’t waste your time trying a formal approach via Lean. (I’m looking at you Dmitry Khanukov) Computational complexity is very messy to formulate technically. I can’t get an AI willing to give me a full Lean-verified proof of something trivial like P closed under complement, forget the PCP theorem. If someone or something does come up with a P ≠ NP, it’ll be following the right intuitive approach, not a formalistic one.

QA harness built for Claude Code | E2E testing with screen recordings, console logs, network HARs, and Playwright traces - wizenheimer/canary

A from-the-ground-up walkthrough of how modern LLMs work, from tokens to transformer blocks to the next-token loop.

Generate hands-on, multi-part technical tutorials on demand, with LLM skills tuned to make content approachable. Then you work through them yourself, by hand.

nice writeup on the bright data sdk that’s being used in mobile apps, specifically TV apps here, that’s doing the web scraping for LLMs