🇺🇸 coalition of anons calling themselves an ai lab. | @elder_plinius, @neurodiverge_, @arXivBangers, Anonterprises | pitches: @basedanarki
❤️s & 🧠s
Joined March 2024
- Tweets946
- Following47
- Followers190
- Likes24.6K
Joywrite retweeted
“3% of all compromises in computing history happened in the last week”
The only way out is through.
They're calling it the Bugpocalypse. LLMs are being used to find long-hidden cracks in the software infrastructure, and it's happening at a furious rate. 3% of all compromises in computing history happened in the last week, claims one source? I believe it.
There are people saying public access to the frontier models ought to be locked out, or heavily guardrailed, over this. The problem with that idea can be summed up in one sentence: when lockpicks are outlawed, only outlaws will have lockpicks.
The lockpicks will be Chinese open-weight models, if nothing else. There's no way to stop those from becoming ubiquitous.
There are two futures in front of us. One in which only the bad guys have lockpicks, and security holes go undiscovered until they're exploited. Another in which everybody has lockpicks and is aggressively red-teaming their own software so it gets fixed before the holes are exploited.
The corpus of human code is huge but finite. The number of defects in it has to be finite too. We can fix those defects now; we have a million friendly robots to do it with.
The only way out is through.
Joywrite retweeted
Try changing what gets carried forward, not just adding compute. This week's experiments span shared caches in looped models, harness-managed agent state, and synthetic priors for learning languages in context.
arxivbangers.substack.com/p/…
Joywrite retweeted
INTRODUCING: Cowcraft MCP and WoWBench - test your AI agent's intelligence in WoW
>point your agent to cowcraft.world/mcp
>agent logs in and plays WoW
>visible in-game so you can watch it play
>WoWBench tracks model performance live
Joywrite retweeted
another 1
INTRODUCING: Cowcraft MCP and WoWBench - test your AI agent's intelligence in WoW
>point your agent to cowcraft.world/mcp
>agent logs in and plays WoW
>visible in-game so you can watch it play
>WoWBench tracks model performance live
Joywrite retweeted
This document is going to be an assigned reading in college classes that cover this moment in time, there's a lot happening in a few paragraphs... ahmath.org/statements
Joywrite retweeted
Test how far your existing models can go without full retraining. This week's recipes explore longer reference chains with tiny LoRAs, editable agent memory and post-trained token drafters.
arxivbangers.substack.com/p/…
Joywrite retweeted
new study reveals i’m better than 4.17 billion other men
based off tenso's list, here's the list of top 50 men on X that are worth following. :)
here's the link: jael.io/50-men
Joywrite retweeted
based off tenso's list, here's the list of top 50 men on X that are worth following. :)
here's the link: jael.io/50-men
based on popular demand, here's the list of top 50 women on X that are worth following.
these accounts are unique, have taste and post bangers that you would rarely see in other accounts.
here's the link: top50-women-x.pages.dev/
Joywrite retweeted
so Anthropic files for IPO and then immediately drops a blog post calling one of their biggest open-weight competitors a danger to the world? classic!
now seems like a good time to point out that I built and run OBLITERATUS, an abliteration framework with over 8k stars and 1.5k forks on GitHub, using Anthropic’s very own models! 🤗
Joywrite retweeted
Try changing how agents improve, not just what they optimize. This week's experiments span pruning self-edits, ordering RL by reward reliability, and checking trajectories for failures that passing scores hide.
arxivbangers.substack.com/p/…
