ai – PwnDefend

one step closer to Skynet?

Yesterday I ran a pentest against an RDP server, the process was ok but not amazing, I had to provide more help than I would have liked, resource consumption and the idea it should keep going…. wasn’t great. The process and output wasn’t terrible at all but it didn’t blow me away.

So today I wanted to see if Claude could take on as simple active directory lab! Now let’s be clear, there were I think one or two updates to the Claude client in that time! The Claude UI even changed look and feel! So I span up an AD lab I had made a while ago and got to work!

AI as the penetration tester

AI hype is everywhere, and don’t get me wrong, I’m a heavy AI user, I’m creating tools, conducting analysis and using AI constantly these days in my work a research. But I wanted to see, how well does an agent work if we try and give it the whole task to conduct with as minimal help as possible.

Threat Intel

FortiSIEM CVE-2025-64155 Exploitation Analysis

‘An improper neutralization of special elements used in an OS command (‘OS Command Injection’) vulnerability [CWE-78] in FortiSIEM may allow an unauthenticated attacker to execute unauthorized code or commands via crafted TCP requests.’

https://www.fortiguard.com/psirt/FG-IR-25-772

This analysis was conducted using data from Defused, enrichment from IPINFO and SHODAN and then analysis using an LLM (GROK) (so take the analysis with a pinch of salt):

Threat Intel

Fortiweb – CVE-2025-58034

‘CVE-2025-58034 is an OS command injection vulnerability (CWE-78) in Fortinet FortiWeb, allowing an authenticated attacker to execute unauthorized code on the system through crafted HTTP requests or CLI commands. It affects versions including FortiWeb 8.0.0-8.0.1, 7.6.0-7.6.5, 7.4.0-7.4.10, 7.2.0-7.2.11, and 7.0.0-7.0.11. The vulnerability has a CVSSv3 score of 6.7 (medium severity) and has been observed exploited in the wild, prompting its addition to CISA’s Known Exploited Vulnerabilities catalog.’

Threat Intel

Fortiweb – CVE-2025-64446

Another day another exploit in the wild it seems! (ok I’m a bit slow to this one). Using Defused Cyber’s Honeypots we have another packet to analyse:

Defense

A brief history of AI being used for Defensive…

There’s lots of talk from some people about how AI is going to destroy the world and needs to be regulated, like some kind of SkyNet has been created… Well I thought, aside from this being largely not reality, I would look into a history of ‘AI’ being used in cyber defence.

Can AI replace intelligence analysts?

Ok, it’s late, and well I wanted to look into cyber attacks where social engineering is a key component combined with technical hacking skills.

There’s been a growing number of these style events, so I tasked GROK to create an assessment for me, let’s see how it did! Let’s both try and answer the questions:

Can GROK replace intelligence officers and can GROK help us defend better against social engineering + technical attacks? What do you think? (please take all of this with a pinch of salt… LLMs are known to make mistakes/hallucinate/lie in a very convincing manner)….

they look nice…. but looks can as we know, be deceiving! (is the entire blog just a social engineering experiment by me?)

News

Using AI to analyse cyber incidents

Currently there appears to be a relatively significant cyber security incident at Marks and Spencer. So I thought I would give a demo of using AI (LLM, GROK) to create a timeline:
(since this is generated by an LLM please take this with a health pinch of salt, I wanted however to show the actual content and method I’ve used so people can understand it)

Education

A Cyber Noir Detective Game

Recently vibe coding has been the name of the game! So whilst dealing with an incident I was thinking about some of the common challenges organisations face when it comes to incident response, which led onto the broader topics of why do so many orgs either have no policies or defined processes but even when they do, people don’t follow them.

So much focus is given to cyber awareness training for ‘end users’ but not so much about training IT and business teams in how to manage incidents.

Enter: Gamified training + comic books + detectives!

Fiction

The Hacker on a Train

It was a crowded train ride during rush hour, and everyone was packed in like sardines. The train was slowly making its way through the city, and people were trying to kill time with their phones or laptops.

At one end of the train, there was a young man, who appeared to be in his mid-20s, typing away furiously on his laptop. He had a serious look on his face, and his eyes were focused on the screen.