Agree on verification, but I'd push it further. Memory safety is a symptom, not the cause. We got here because of design decisions and trade-offs we made all along, and you can see it in pretty much every recent Linux kernel disclosure. Throwing more tokens at the attack surface won't save us either. Nothing we have today is ready for AI-driven vuln research at this scale, and finding and fixing bugs with AI doesn't change that. What we actually need is redesign and architecture validation with AI, formal methods where we can, and not accepting attack surface at the design phase in the first place. That's just not how we build software today. Patching known bugs doesn't touch the root cause, and hardware has the same problem.

I started learning low-level software when reading and implementing graphics engines. In the mid 90s, you wrote hand-optimized assembly to render via the CPU. The first x86 assembly I ever wrote was an anti-aliased scanline renderer. Analyzing this bug was blast from the past.

Introducing Claude Sonnet 5.5, the second model in the Claude 5.5 family. It’s a clear upgrade over Sonnet 5, runs more than 30% faster, and costs up to 30% less for most work.

This topic is more complicated than it looks. As an industry, we've definitely made a lot of progress in cyber over the years. But this is different, very different from what we've dealt with before. Look at the spike in fixed vulnerabilities across all the big vendors. And this isn’t a one-off event, it's a continuous trend. What it shows is that our understanding of cybersecurity in general is right, but the way we've solved these problems in the past is not. AI is exposing all the obscurity and complexity across the tech stack, including hardware, in exactly the places where problems were traditionally hard for human researchers to find. At the same time, threat actors are getting a force multiplier from AI without investing in new resources or building new expertise. And on top of that, every new model outpaces the previous one, so yesterday's statements and assumptions quickly become wrong. This isn't a static risk. It's highly dynamic, and things are changing and escalating far faster than the industry, or the tech stack, can react.

Our CFP is open and we're especially looking for hacking magic 👾🪩🪄 Learn more about the content we're looking for on our Sessionize! sessionize.com/districtcon

Another point, the HW and FW security boundaries are our last frontier for building real hardwired boundaries for virtualization, memory isolation, and hardened agent sandboxing. In the end, the whole confidential computing movement is more about cloud cost optimization than anything else. It's just broken by design, you can't build a secure stack if what's underneath has been broken for many years.

Our craziest escape yet: The @Accomplish_ai research team was able to exploit a vulnerability in Cloudflare Containers that let a sandbox read other customers' files - SQLite DBs, Chromium profiles, .env files etc, Cloudflare Sandboxes and Browser Run run on the same disk implementation and were affected too. We reported this to @Cloudflare, who super quickly fixed it. Read @CloudflareDev post in collaboration with Accomplish researcher @orenyomtov on their official blog: blog.cloudflare.com/containe…

As Flare-On starts I am happy to announce the official Hex-Rays IDA MCP Server is out! 🥳 Details and links below ⬇️

I'm hiring an exceptional Offensive Security Researcher for my team at NVIDIA (Offensive Security Research - OSR). Firmware, microcode, RISC-V, hypervisors, and shipping mitigations like HW CFI, Memory Tagging, and Pointer Masking from the ground up. jobs.nvidia.com/careers?quer…

I put my @UnpromptedAU slides up at justdionysus.github.io/slide… — a bit of reflection on exploit development in the age of AI. My TL;DR is keep pushing to understand complex things, be honest with your own understanding, and use AI as a power tool to increase pace and depth.

Introducing Claude Opus 5.5, the first model in our new Claude 5.5 family. It performs at the level of Claude Fable 5.1 for most tasks, and costs 40% less to run than Opus 5.

We implemented 1024-bit RSA signature forgery in nearly SNFS time! Temporary access to an HSM allows an attacker to forge arbitrary signatures (without factoring). Join work with Laura (1st author!), Adam, Nadia, and Emmanuel github.com/ucsd-hacc/NSNFSSS…

The era of AI requires a significant paradigm shift in the hardware and firmware layers. Today we are building on top of hardware and computational primitives that were defined decades ago, and they don’t serve us well for the future of AI infrastructure at the current exponential scale. If you look back at history, it’s always outdated patterns that hold us back from progress. Every transformational shift happens when we invent a new paradigm that defines the next breakthrough in technology evolution. I’m very excited about it, and it’s time to build an AI-native hardware ecosystem.

Your weekend reading assignment has arrived early. A first taste of the upcoming, still-under-wraps Phrack 73: “THE PROXY THAT MADE NO SENSE” by @mikko. archives.phrack.org/dl/73/th…

The slides from my talk at Microsoft Bluehat Singapore are public here: thomasdullien.github.io/abou… It's my first BlueHat talk since the Vista days.

This tweet has aged a few months, and frontier progress has only strengthened my feelings in this direction. Twelve months ago my thought was that AI is just another tool, automating parts of the RE/VR process. Now I'm more of the impression that it operates like the creative mind of an experienced researcher, who thinks beyond established patterns and finds new attack paths. The most effective vulnerability researcher is no longer human.

The @reconmtl talk "SELECT * FROM binary - Vibe Reverse Engineering with IDA, Ghidra and Binary Ninja" is out. piped.video/watch?v=ADPoD7-_…

We've reached the moment in time where (unsafeguarded, unmonitored) AI actually does just pose a national security risk. The biological misuse we caught is the most concerning to me. We work hard to stop this. But in a world of proliferation, we need to rapidly build defenses against it. (I'm actually fairly optimistic about biodefense + cyberdefense) This is an incredible megareport by our threat intel team