
Me explaining to AI dudes that cybersecurity ≠ vulnerability management and that the majority of cyberattacks involve exactly zero vulnerabilities
Project Glasswing: * 37,966 findings. * 2 exploited in the wild. vulncheck.com/blog/anthropic…

Security Researcher | Cloud Security | Speaker

Me explaining to AI dudes that cybersecurity ≠ vulnerability management and that the majority of cyberattacks involve exactly zero vulnerabilities
Project Glasswing: * 37,966 findings. * 2 exploited in the wild. vulncheck.com/blog/anthropic…

Amazing breakdown regarding the current state of deception and whether it will finally transition to a "must-have". I'm optimistic that we can finally overcome that hurdle and begin implementing deception for organizations of all sizes.
Decoys can reveal intruders and AI agents that exceed their scope. Will the renewed interest in cyber deception leave a lasting mark on how we defend?

We're definitely going to see a rebalancing of organizational tolerance for attack disruption. Will be a necessary component in order to deal with "agentic swarm" attacks
ARTEX VS MDE! not gonna lie! MDE has kicked it's ass!




This 100%
Deception is more important now than ever before, thanks to AI. Deception can go where traditional detections are not feasible or realistic.


This looks like an awesome resource! Definitely check it out
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-…


Very interesting
TIL: Microsoft Copilot keeps conversation history and personalization memory, stored in the user’s Exchange mailbox, so retention and eDiscovery apply. Prompt history, data artifacts, and memory data are stored in hidden folders in the user's Exchange online mailbox, similar to Teams messages. This means that an attacker with full access to mailboxes can pull this information. techcommunity.microsoft.com/…

Completely agree here - You should patch PHP but the sky is not falling yet. The issue here is going to be all the IOT Crap that just does not ever get updated. There still is time to patch. If you want to see the state of things, find a cyber range (I prefer HackTheBox). Go look for a bunch of old PHP Boxes and try the exploit out on them. Right now I think you are going to have a tough time finding a retired box that you can exploit with PHP2SHELL.... Atleast I hope you do because I am having a hell of a time doing it and both love/hate to learn how much of a noob I am at times. AI is really good at making crap look good. I know the POC that shows a POC of <?php echo("OK") ?> being exploited sounds scary. But when you look deeper there are a lot of hard-coded offsets that make it only work on that target.
I'll be the first to admit I'm mediocre at exploit development, but I am in contact with enough extremely talented exploit developers to say this for certain: Anyone claiming they were "easily" able to generate a working php2shell exploit with AI is either: a) a world-class exploit developer with an extreme case of imposter syndrome. b) currently looking at some very confident-sounding but wrong LLM output. I'm rather certain it's the latter based on the number of AI-hallucinated exploits I've seen posted on GitHub. Threat actors are also blasting my honeypot network with "php2shell" exploits that not only don't work, but aren't even within 10,000 miles of anything to do with how the exploit actually works. I've seen the same technique from the attached image in multiple alleged "php2shell exploits", including one on GitHub with over 100 stars. Problem is, it's not php2shell, nor is it even a vulnerability at all. It's a technique known as "session poisoning", which can be leveraged to exploit local file inclusion (LFI) vulnerabilities in PHP apps (not PHP itself). The only thing it has in common with CVE-2026-109618/CVE-2026-109619 is they both use the "PHP_SESSION_UPLOAD_PROGRESS" feature. And mitigation advice for both issues reference setting "session.upload_progress.enabled" to "Off". People are shoving loose details of the vulnerability into LLMs, the LLMs are keying in on the "PHP_SESSION_UPLOAD_PROGRESS" string, then generating session poisoning code that has nothing to do with the actual CVE. And because the people doing this have absolutely no idea what they're doing, they're confident enough to post their "proof of concept exploit" to GitHub and share it on social media. So if you see people talking about "session poisoning + LFI" it's likely their LLM hallucinated an exploit based on a 2018 blog by Orange Tsai talking about a local file inclusion attack vector. What's even worse is that "security researchers" are then using LLMs to triage these exploitation attempts, falling victim to the same hallucinations, then confirming there is ongoing mass-exploitation (there currently isn't). The actual real exploit is very difficult to exploit, and so far I've not seen anyone successfully get it working on more than a handful of PHP version with highly-specific configurations. LLMs are useful tools in the hands of skilled professionals, but for others, they're often just a Dunning-Kruger accelerator.

Has to be one of the most diabolical phishing tests ever
What the actual fuck have I just read

Starting on November 4th, Entra ID users will be able to create agent identity blueprints unless this capability is disabled. I strongly recommend you disable this capability so you don't have users creating agent identity blueprints. There's a setting to turn this off in Entra ID under User Settings: "Users can register agent blueprints". Switch it to "No" to prevent users from being able to do this. Agent identity blueprints are templates that define identity configuration and credentials for AI agents. This is similar in capability to app registrations but for agent identities. Allowing all users to create blueprints has similar risk to allowing all users to register applications (default is enabled as well). When a user creates an agent identity blueprint, they become the owner of it until removed. This provides full control access to the blueprint. Agent ID Developer and Agent ID Administrator will still have the ability to create them. More details here (including searching for existing blueprints): lazyadmin.nl/office-365/micr…


Again?!
Not a professional lip reader but did Dan Quinn just say/ask “That was fourth down?”