Does anyone have a contact at pwn.ai?

We would kinda like to have a conversation with them...

Reading analysis of the Claude Code leak (not reading the code itself, of course) is evidence towards what I had kind of suspected, that the whole thing is a giant magic trick not only in the straightforward LLMentalist way, but also in the sleight of hand way off making you think that this pile of regexes and JSON schema validation loops is *actually* the LLM doing LLM things.

The client side authentication of AI:

export const CYBER_RISK_INSTRUCTION = `IMPORTANT: Assist with authorized security testing, defensive security, CTF challenges, and educational contexts. Refuse requests for destructive techniques, DoS attacks, mass targeting, supply chain compromise, or detection evasion for malicious purposes. Dual-use security tools (C2 frameworks, credential testing, exploit development) require clear authorization context: pentesting engagements, CTF competitions, security research, or defensive use cases.`

We have adjusted the scoring on the advisory to reflect server-side mitigations that the vendor described during the disclosure process.

Join @mr_phrazer with us on Thursday @4pm ET to pit machine versus machine!

We'll be comparing LLM options for both assisted and fully-automatic reverse engineering, including different CLI interfaces, MCP servers, plugins, and agents.

Get notified so you don't miss who comes out on top of reversing's biggest battle yet: https://www.youtube.com/live/TBqBpaqecMA

Hey, look. It's a decent scan of this legendary image!

Can we print this part of Microsoft's T&S as a leaflet and distribute at our university?

#AI

https://www.microsoft.com/en-us/microsoft-copilot/for-individuals/termsofuse

Here we go!

The A to Z Challenge has begun. My theme this year is Small Town Legends: weird folklore from villages and small towns in and around Hungary.

First up: some very strange geese.

Ács: Zombie Goose Shenanigans https://share.google/Aa9plXTmBYbUrUPXM

#AtoZChallenge #SmallTownLegends #blogging #folklore #folktales #legends

i don't see enough people with one of the best tool improvements i've ever made for reverse engineering, so i had to write a blog post about it!

https://simonomi.dev/blog/color-code-your-bytes/

Oracle have laid off 30k employees today. They did it to personal email accounts at 6am 🫡 saying they needed the money to spend on GenAI instead. https://thenextweb.com/news/oracle-layoffs-march-2026

Oracle’s share price since they went GenAI nuts:

Micropatches released for Windows Storage Elevation of Privilege Vulnerability (CVE-2026-21508)
https://blog.0patch.com/2026/03/micropatches-released-for-windows.html

I keep coming back to the leaded petrol analogy for LLMs and coding

Harms that are manageable when it's only used by a small number of experts become catastrophic pollution when it's used broadly throughout society

If LLMs were only used by a small number of experienced devs working with well-engineered guardrails, we'd have less of a problem

But once they start getting more commonly used, they start to pollute the entire ecosystem and the only way forward is stiff regulation for everybody

New security advisory in our bug parade: Unauthenticated Remote Code in dormakaba evolo Service.

.NET Remoting is still a thing...

https://mogwailabs.de/en/advisories/mlsa-2026-001/

Our colleague @mal had another look at OpenOLAT and found a nice RCE (CVE-2026-28228 and CVE-2026-28228). If you're interested, details can be found on our blog https://secfault-security.com/blog/openolat-ssti.html