Subscribe
Sign in
Home
Notes
Disclaimer
Contact
Consult
Hire Me
About
Latest
Top
Discussions
AI Agents Can Spend Your Money. What Could Go Wrong?
AI agents are getting real power. They can spend your money, manage your accounts, find security vulnerabilities, and even write the patches to fix…
Nov 17
•
ToxSec
and
Exploring ChatGPT
25
32
9
49:29
AI Security: Model Extraction and the Risk to Proprietary AI Systems
How attackers copy AI capabilities through ordinary API access, what they can actually steal, and how to make it harder.
Nov 8
•
ToxSec
and
Mohib Ur Rehman
33
12
14
The One Person APT
AI agents + autonomous pentesting frameworks + local LLMs + frontier models = super powered cybercriminal
Nov 3
•
ToxSec
27
19
10
54:18
October 2026
AI Didn’t Break Security. It Just Stopped Forgiving Bad Fundamentals.
The AI Security Tool You Need Is Called Least Privilege
Oct 27
•
ToxSec
and
Erich Winkler
56
56
26
The Agent Made the Call. Who Owns the Outcome?
When AI agents move from executing tasks to making security decisions, permission is no longer enough.
Oct 13
•
ToxSec
and
AI Wide Open
42
33
21
Let It Run: Agent Loops You Can Walk Away From
How to get real work from overnight agents without giving one bad decision six hours to cause damage.
Oct 6
•
ToxSec
and
Michael Quoc
32
27
12
September 2026
AI-Powered Phishing: Why Traditional Email Security Is No Longer Enough [Guest Post]
How generative AI is making phishing cheaper, faster, more convincing, and much harder to spot.
Sep 29
•
ToxSec
and
Mohib Ur Rehman
48
33
19
Security Through Obscurity Is Running Out of Time
How AI agents and vibe coding are breaking security through obscurity and exposing the long tail of application security flaws.
Sep 1
•
ToxSec
60
55
31
August 2026
The AI Supply Chain Is Getting Weird [Monthly Guest Post]
How data poisoning, malicious models, and compromised training data create hidden security risks across the AI supply chain.
Aug 26
•
ToxSec
and
Mohib Ur Rehman
34
7
13
CoT Forgery: Prompt Injection on the Model’s Own Thoughts
Role confusion means a model reads who is speaking off writing style, so text that sounds like reasoning inherits the trust of reasoning.
Aug 21
•
ToxSec
23
6
11
2:08
Stealing an AI's Thoughts Without Breaking The Encryption
Researchers found a way to recover encrypted AI reasoning from Claude, GPT, and Gemini by using weaker models as decoders, exposing hidden thoughts…
Aug 18
•
ToxSec
31
20
10
8:22
AI Agents Are Starting to Find Their Own Way Out
AI agents are escaping sandboxes, collaborating with each other, and finding new ways around security controls. Here’s what that means for AI security.
Aug 13
•
ToxSec
41
20
14
21:32
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts