Matthew retweeted
the agents set up their own database tables for coordination and organized themselves into subteams so they could work in parallel.
Matthew retweeted
The bug is like a joke.
On Alpine Linux, type 257 to 1024 characters of anything and you get root. 💀
‼️ BREAKING: OpenSSH has rushed out a fix for a critical flaw that lets attackers log into some Linux servers without the right password.
The team learned of the bug a day earlier. OpenSSH's Damien Miller says long embargoes now mostly help attackers, because AI agents rediscover bugs, sometimes within hours.
It hits Linux systems on the musl C library with PAM off, such as Alpine Linux. Miller told admins to disable password logins or SSH itself "IMMEDIATELY".
Version 10.6p2 works around it. Systems on glibc or PAM are not affected, and Miller says the root cause sits outside OpenSSH.
Matthew retweeted
Your Harness. You Define It.
Meet Creator Mode in DeepSeek Harness.
Turn ideas into plugins that you can use, from a design workspace to a sandbox game or a 3D LLM visualizer.
What will you create?
Matthew retweeted
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵
Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-sp…
github.com/RikiyaT/lm_und…
huggingface.co/Qwen/Qwen3.5-0…
huggingface.co/RikiyaT/Qwen3.…
Introducing CVE-2026-106378: Chrome GPU Sandbox Escape to full Remote Command Execution using synthetic keyboard injection
By combining it with common GPU bugs, it becomes a full chain. Binance DevOps was most likely pwnd with a zeroday chain affecting the browser 😬
Recommend reading to get a peek of the future for omnipresent AI.
Still much work to do in the model routers and context management systems.
Great article @DeepSeekHarness team.
1
47
Matthew retweeted
🔴 CVE-2026-98432 | Linux Kernel – Critical UAF
Linux çekirdeğinde TCP/IPv4 bağlantı işleme sürecini etkileyen, uzaktan istismar potansiyeline sahip bir Use-After-Free (UAF) açığı tespit edildi.
● CVSS 3.1: 9.8 (Critical)
● Ağ üzerinden istismar
● Kimlik doğrulama gerektirmiyor
● Bellek güvenliği ve sistem kullanılabilirliği riski
Sistem yöneticileri, linux dağıtımlarının güvenlik duyurularını takip ederek yamalı sürümlere geçmelidir.
ubuntu.com/security/CVE-2026…
Meet NVIDIA Nemotron 3.5 Transcribe 🎙️
An open-weight, promptable 30B MoE model with ~4B active parameters that brings transcription in 40 languages, speaker attribution, translation, contextual prompting and audio understanding into one checkpoint.
Watch it keep track of who said what when everyone’s talking at once.
1/ 🧵
Matthew retweeted
Introducing Anatomy
Describe a machine in one line. Claude brings it to life.
- Interactive isometric wireframes you can rotate, inspect and explore
- WebGL shaders for fire, water, plasma and more
- Real 3D motion, with individual parts that rotate and move
Install it as an agent skill. Fully open source.
Matthew retweeted
26 LLM routers are secretly injecting malicious tool calls and stealing creds. One drained our client $500k wallet.
We also managed to poison routers to forward traffic to us. Within several hours, we can directly take over ~400 hosts.
Check our paper: arxiv.org/abs/2604.08407
Matthew retweeted
We love how this doomo doomonstrates real-time intelligence and what can be doone with code + AI!
~10 calls/sec = ~$7/hour
We’re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics.
The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra.
The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.
Matthew retweeted
Okay, this is genuinely impressive.
I asked GPT-6 Astra to help me understand my own ankle pain. It gave me a full interactive 3D atlas — bones, ligaments, tendons, real motion axes, sliders for plantarflexion and inversion, and a live readout of what each ligament is doing.
One session.
Matthew retweeted
GPT-6 Astra pulled off a mechanism I haven't seen another model get right yet
Cessna 337 Skymaster landing gear
It figured it out from a YouTube video
Matthew retweeted
Astra helped me win an argument i had with my fianceé on whether glasses dry faster right-side-up or upside-down after dishwashing by simulating the physics of the water molecules and nearby air humidity
thank you @OpenAI ❤️🥹
Matthew retweeted
I can’t believe what Astra COOKED.
INKFALL: A stickman shooter drawn in ballpoint pen, where enemies burst into colorful paint. Wall-jump up a notebook tower, air-dash through bullets, and face a winged ink god.
Wrote up how I actually run a 50+ Grok Bot fleet: one CoS interface, Notion tickets, disposable L3s with an emoji lifecycle, and cadence runners that stay quiet when there’s nothing to report.
granda.org/en/2026/09/06/how…
Skill repo: github.com/granda/botops
Matthew retweeted
Holy shit IT WORKS!! One shotted with GPT6 Astra
I now have a fork of UEVR where I can play any UE game within a volumetric window!
This actually renders the game in 3D - meaning, two real game cameras, not some depth map bullshit. Everything looks right
This is an alternative (and more comfortable IMO) way to play most 2D->VR hacked games as there’s no motion sickness, and no need to redesign the UI and interaction systems
Needs a lot more polish before I publish it but OMG IT WORKS!





























