Memory Safety Bugs in Rust Standard Library
forge-blog.github.io/posts/r…
157
MicrosVuln retweeted
Today @_rybaz and I are publicly releasing all the materials for our Introduction to Harness Engineering for Vulnerability Research training we gave at @unpromptedconf last week.
Our goal was to teach the concepts we've been using to do much deeper security research and make both vulnerability analysis and red team engagements far more impactful.
The workshop can be found here: grahamhelton.github.io/harne…
The slides can be found here: grahamhelton.com/talks/2026-…
Any loose of interest in offensive security due to the strong capabilities of frontier AI models in this area?
1
1
262
MicrosVuln retweeted
🚨NEW INVESTIGATION: zero-click mercenary spyware hacking via targeted ads.
A normal-looking ad loads. If you're a target, you're infected. No click needed.
To protect yourself, block ads now.
First time it's been confirmed in the wild. We call it #AdFection.
Candiru spyware infections at @AJEnglish. 45+ devices infected.
Same ad pipeline operated by Israeli company Discover-Tech reached NATO, European Parliament, dozens of gov networks & critical infrastructure incl US nuclear giant Westinghouse.
Confiant, which worked with us, found signs ads were displayed at US @DeptofWar, IAEA & ICC.
Investigation ongoing 1/
READ our report by @billmarczak @WolfieChristl, me Astrid Perry, Siena Anstis & Ron Deibert : citizenlab.ca/research/how-d…
Confiant's blog: blog.confiant.com/p/adint-we…
MicrosVuln retweeted
LLMs are great at finding bugs, but so are fuzzers. @5aelo compared the vulnerabilities found in V8 by BigSleep, an agentic system, with those found by the Fuzzilli fuzzer.
projectzero.google/2026/10/f…
MicrosVuln retweeted
JUST IN: Abliterated DeepSeek-V4.1-Flash @ $0.20 IN / $0.80 OUT, plus enhanced with cyber RL for exploit development. Now 88.1% on CyberGym. Match Daybreak performance with no approval required.
Get $10 on us while supplies last
MicrosVuln retweeted
(2) Any dislike of AI-assisted bug reports is out of place. It shouldn't matter whether an AI or a guided fuzzer found a bug; both are computational searches. Unfortunately, there are still people that reject bug reports accompanied by PoCs based on the origin in AI.
MicrosVuln retweeted
Europe is the only region stupid enough to invest all it's money into other region's economies and then complain that their own economy doesn't improve.
🇺🇸🇪🇺 Europe once had factories and no money. Now it has the money, but it sends much of it away.
Where do European savings go? Across the Atlantic, into America, where they flow into US equities, Treasuries, and other financial assets—and, increasingly, into the valuations of a handful of tech giants.
Those returns are difficult for American factories to match, which helps push capital toward finance and technology rather than manufacturing.
Meanwhile, European plants face high energy costs, weak investment, and growing Chinese competition—all problems that more capital could help address, yet European savings continue to finance American assets.
I call this “the West’s carry trade against itself”: Europe saves; America turns those savings into financial assets and inflated tech valuations; meanwhile, China builds the factories.
The chart below shows twenty years of this dynamic.
More on this soon: open.substack.com/pub/europe…
This seems to be a known issue with @OpenAI Cyber verification, but no one appears to be addressing it.
Has anyone else been stuck on “Couldn’t verify photos” even after trying multiple devices and browsers?
@ChatGPTapp @OpenAIDevs @OpenAINewsroom
2
223
a Tsinghua University lab just put a project on GitHub that replaces a $400,000 H100 rack with a single 24GB graphics card.
it's called ktransformers, and the trick is almost stupidly simple: the experts you actually use stay on the gpu, the ones you don't sit on the cpu until they're called
/ deepseek-v3 and r1 with 139K context in 24gb of vram
/ up to 28x speedup over the standard setup
/ fine-tune deepseek-v3 across four rtx 4090s instead of a datacenter
/ built by tsinghua university's madsys lab, not a startup with a landing page
apache 2.0, and already past 17,000 stars.
-> github.com/kvcache-ai/ktrans…
bookmark it.
MicrosVuln retweeted
By reverse engineering AirDrop and building the AIRFUZZ fuzzer, @microsvuln @notippenhauer uncover six vulnerabilities across macOS, iOS, Android, & Windows, incl. zero-click DoS bugs, authentication and encryption bypasses, and use-after-free with remote-code-execution impact.
Anyone knows that if there is any way to use Claude's Fable to do cyber security activities? (Even with verified organization for cyber security activities) ?
@moyix Maybe you know ?
1
1
323
MicrosVuln retweeted
The eighth article of the Exploiting Reversing Series (ERS) is now live. Titled “Exploitation Techniques | CVE-2024-30085 (Part 02)” this 91-page technical guide offers a comprehensive roadmap for vulnerability exploitation:
exploitreversing.com/2026/03…
Key features of this edition:
[+] Dual Exploit Strategies: Two distinct exploit versions leveraging the I/O Ring mechanism.
[+] Exploit ALPC + WNF OOB + Pipe Attributes + I/O Ring: elevation of privilege of a regular user to SYSTEM.
[+] Replaced ALPC one-shot write with Pipe Attribute spray for I/O Ring RegBuffers corruption: more reliable adjacency control.
[+] Exploit WNF OOB + I/O Ring Read/Write: elevation of privilege of a regular user to SYSTEM.
[+] Pure I/O Ring primitive: eliminated ALPC dependency entirely. WNF overflow directly corrupts I/O Ring RegBuffers for arbitrary kernel read/write.
[+] Solid Reliability: Two complete, stable exploits, including an improved cleanup stage.
This article guides you through two additional techniques for exploiting the CVE-2024-30085 Heap Buffer Overflow. While demonstrated here, these methods can be adapted as exploitation techniques for many other kernel targets.
I would like to thank Ilfak Guilfanov (@ilfak ) and Hex-Rays SA (@HexRaysSA ) for their constant and uninterrupted support, which has been vital in helping me produce this series.
I hope this serves as a definitive resource for your research. If you find it helpful, please feel free to share it or reach out with your feedback!
Enjoy the read and have an excellent day.
#exploit #exploitdevelopment #windows #exploitation #vulnerability #minifilterdriver #kernel #heapoverflow #ioring
به تیم توسعه پیشنهاد میکنم در وبسایت رسمی و گیت هاب آدرس ایمیلی معرفی بشه که متخصصان امنیتی که مستقلاً قصد تست امنیت نرم افزار رو دارن تست کنن و در روند ایمن سازی «زمزمه» بتونن با سرعت بالا با تیم توسعه همکاری داشته باشن. متاسفانه اپلیکیشن در حال حاضر صد در صد امن نیست .
معرفی پیامرسان کاملا امن «زمزمه» برای حفظ ارتباط با یکدیگر در زمان قطع اینترنت
پیامرسان «زمزمه» که توسط گروهی از برنامهنویسان آزادیخواه و مورد اعتماد طراحی و ساخته شده است، در حال حاضر فقط برای تلفنهای اندروید و به صورت فایل APK در دسترس است، و به زودی بر روی گوگلپلی نیز قرار میگیرد.
هر چه عده بیشتری این پیامرسان را نصب کنند، کارآیی آن در زمان قطع اینترنت بالاتر خواهد بود. بنابراین توصیه میکنیم که زمزمه را هماکنون نصب و در همه گروههای خانوادگی و دوستی نیز تبلیغ کنید.
لینک دانلود: zemzeme.app/download.html
MicrosVuln retweeted
I am pleased to announce the publication of the sixth article in the Exploiting Reversing Series (ERS).
Titled "A Deep Dive Into Exploiting a Minifilter Driver (N-day)", this 251-page article provides a comprehensive look at a past vulnerability in a mini-filter driver:
exploitreversing.com/2026/02…
It guides readers through the entire investigation process—beginning with binary diffing and moving through reverse engineering, deep analysis and proof-of-concept stages into full exploit development.
I hope this serves as a valuable resource for your research. If you enjoy the content, please feel free to share it or reach out with feedback.
Have an excellent day!
MicrosVuln retweeted
Lots of people are claiming this is useless, unimpressive, etc but I will note that compiling the Linux kernel on x86_64 to the point where it could boot and run took Clang/LLVM about 10 years! And they, too, had access to a reference gcc compiler :p
New Engineering blog: We tasked Opus 4.6 using agent teams to build a C compiler. Then we (mostly) walked away. Two weeks later, it worked on the Linux kernel.
Here's what it taught us about the future of autonomous software development.
Read more: anthropic.com/engineering/bu…



























