My research got accepted into Phrack!
In it, I explain the Windows Smart Card Minidriver architecture and show a really cool piece of malware I made that can use any smart card connected to a remote computer, regardless of model.
https://t.co/BFpGDI6S0m
Researchers(Mark Ermolov, Dmitry Sklyarov, and Maxim Goryachy (from Positive Technologies)) found two undocumented x86 instructions, udbgrd and udbgwr, that can read and write microarchitectural data directly.
They enable access to the microcode patch RAM and match-and-patch structures inside the CPU. UDBGWR is capable of modifying MSRAM and updating match-and-patch entries that override target instructions. Intel has never documented these instructions but they're callable from user space and they give direct write access to the CPU's microcode storage.
PoC for CVE-2026-111497 is now public with full source code. The Brotli out-of-bounds write in Apple’s libcompression that was patched in iOS 27.2.1.
Reproduced on iOS 27.0.1 through the system network stack: a server sends a Content-Encoding: br response with a two-stage Brotli stream. Stage 1 parks the decoder waiting for more input. Stage 2 corrupts the output position tracker, and the backward-reference copy writes outside the ring buffer with attacker-controlled data and length.
Root cause: the decoder accepts BlockSwitch commands even when fewer than two block types exist. Upstream Brotli 1.2.1 rejects them with error code -17. Apple’s fix in 27.2.1 is an exact backport of that commit.
iOS 27.0.1: vulnerable. iOS 27.2: vulnerable. iOS 27.2.1: fixed. iOS 27.3 beta 1: still vulnerable.
The repo includes trigger generator, decode harness, sample binaries, and the instruction-level diff of Apple’s patch between 24B119 and 24B156.
https://t.co/XoVEg4MnPG
The ARM build of macOS 27.2 now boots on ordinary Intel and AMD PCs.
Stela v0.1.0 is out today. This isn’t a Hackintosh build: it takes the unmodified Apple silicon version of macOS and runs it on x86, translating arm64 code to x86-64 as it goes, with its own Metal driver for NVIDIA cards. The screenshot is a Ryzen 7 9800X3D with an RTX 5080.
To macOS, your PC looks like an M2 Mac mini. macOS and most apps start.
What you need:
An Intel or AMD CPU from about 2011 or newer, 8 GB of RAM, Secure Boot off
An NVIDIA GTX 16 or RTX 20 series card or newer
A USB stick (8 GB or more) and an empty 60 GB partition
Network during the install, since it downloads macOS for you
Setup runs from Windows, Linux or macOS and builds the boot USB. Open source, written in Rust.
https://t.co/BuaqlAMWLg
The bull case for formal verification is "Here is a reusable library of properties/specs you can import in your code" not "Here is a few million LOC of Lean for you to go over, but the model passes!"
If you remove different representations, you find out that many algorithms are either the same, or almost the same.
Shortest path is semiring based computation.
More here: https://t.co/IiKh56Cijg
I realized this when I was writing my book on OR algorighms. There are hundreds of algos I went through. Sometimes, you could find 40-50 algorithms, that are different only in one parameter, or one function in the function composition.
---
I am 100% sure, there are abstract algorithms in space search + reduction, that could simplify what already exist and provide dozens of real world useful algos for thousands of various usecases.
New research from Microsoft Research demonstrates that reprogramming just a few dormant neurons in an AI agent can create a backdoor that leaves the agent behaving normally until it hits a specific trigger.
While the specifics of this threat are novel, the mitigation is not: Know where your model came from, and be smart about its permissions. Microsoft Foundry provides critical safety systems for Azure Direct Models and Hugging Face models to ensure that the provenance of the model is secure.
Microsoft researcher @DimitrisPapail explains the measures that model hubs and vendors should take as well as mitigation strategies for enterprises and users. https://t.co/2iTKdX03Xz
man, who should i trust? a bunch of nerds who made linux run on apple silicon, reverse engineering that pesky gpu isa among a gazillion other things, all pre-ai.
or the guy who thinks having your agent rewrite shitty rust he never looked at in x86_64 is excellence?
Two thoughts on this:
1. Very cool. As someone who's primary interest is the offensive side of cyber, I love it. *chefs kiss* and congrats to the team.
2. You cannot bug hunt your way to 'secure'. AI doesn't change that. If your strategy for the next few years is to boil the oceans continuously scanning your same codebase, with the same broken architecture, with the same broken threat models, you are going to have a bad time. a) you're going to spend a fortune and b) you're going to keep getting hacked.
Are you a fellow PL nerd, early-ish in career? Interested in spreading the gospel of fancy types and strong gurantees? Looking for a job? Come work with me in the MSVC Rust Compiler (+ sanitizers) team, we're hiring!
https://t.co/7VqN6LPnQV
eMudhra key leaks impact Android's Factory Reset Protection (FRP), thanks to the generous GMS sign-in interface.
TL;DR: Inject mm.setContentFilters('{"user_setup_complete":1, "device_provisioned":1}'); on server JS responses with `mitmproxy --emudhra` and reboot, bypassing SUW. Then, flip system_server's mFrpActive with any N-day to leak the secret and wipe FRP.
Demo below. Full chain private for obvious reasons.
i thought read() was a syscall wrapper, but it can dlopen a bytecode interpreter and execute your code
"house of windy" turns this into leakless code execution with only OOB null byte writes
https://t.co/Rvzd6glGv3
https://t.co/gOtQb9QmuU
"At present, it appears that highly-automated, LLM-
based remediation pipelines are more likely to change application behavior, introduce new vulnerabilities, or mask existing ones, than fix known vulnerabilities."