🤦 I mean... at least they could have had an old win 2000 server or so? denmark's national registry was "breached" and exfiltrated due to the password '123456'.

https://www.tomshardware.com/tech-industry/cyber-security/hacker-claims-123456-password-let-them-breach-the-danish-national-registry-accessing-8-8-million-records-it-and-data-services-company-with-weak-passwords-linked-source-of-the-hack

It looks like whatever you try, the containment might be just temporary, agents successful escape their environments, using knowledge or finding new 0-days ¹

It's getting too cheap to run attacks on software, a recent report of automated agentic hacking discovers at least 600k stolen credit cards, from the report I'd guess it was a Chinese speaking people, but who knows. Mitigation seems hard considering the scale and experience modern ai harness can bring into such campaign. SQL filtering, web application firewalling, monitoring what's going on on the system, filtering filesystem status (e.g. are there scripts which should not be there?) might be helpful. 🙈

https://gambit.security/blog-posts/autonomous-ai-agents-online-retailers-25-a-company

First reports coming in that ai is fed with wrong information by attackers trying to trick users into going to wrong websites or calling fake numbers. Quiet interesting in context of these Google search, ai answer:

https://medium.com/@arielsimon/dark-sourcery-how-hackers-manipulate-ai-to-scam-you-88df434d2073

I think we just see the first glimpses of what will be happening the next years to come, with increasing speeds you will see devices getting owned and manipulated on a scale never seen. It might not be "skynet" taking over everything in the beginning, but for sure humans will easily.

Someone hacked his webcam, mic, hdmi capture card, lightning, monitor using claude and prompting: https://schlarp.com/posts/everything-i-own-owned/

an interesting talk about how agents hacked internally around, build a message board for coordination and all without humans interacting.