Sign in Subscribe

AI Security

AI Security explores the risks, architectures, controls, and governance required to secure artificial intelligence across the enterprise. Coverage includes generative and agentic AI, LLMs, models, data, identities, tools, MCP, authorization, supply chains, attacks, and emerging security architectures—with a focus on what CISOs need to understand, govern, and defend.
02
Oct
Part 9 — Security Infrastructure Can Become Offensive Infrastructure

Part 9 — Security Infrastructure Can Become Offensive Infrastructure

Security tools increasingly consume hostile data, interpret it with AI and act with privileged authority. DEF CON 34 showed how telemetry, evidence and agentic security workflows can be manipulated—turning the infrastructure built to defend the enterprise into an attack path.
12 min read
02
Oct
The Shadow AI Economy: From Shadow IT to Shadow Agency

The Shadow AI Economy: From Shadow IT to Shadow Agency

Shadow AI is becoming Shadow Agency. At PAC Horizons 2026, I discussed how AI agents, identities, tokens and delegated authority are changing the CISO agenda — and why Least Privilege may no longer be enough.
8 min read
11
Sep
The Automation Paradox: AI Can Fix Faster — But Can We Trust the Fix?

The Automation Paradox: AI Can Fix Faster — But Can We Trust the Fix?

17 min read
11
Sep
AI Is Changing the Economics of Vulnerability Discovery

AI Is Changing the Economics of Vulnerability Discovery

16 min read
11
Sep
Prompt Injection Is Becoming an Execution Problem

Prompt Injection Is Becoming an Execution Problem

14 min read
10
Sep
When Security Tools Become the Attack Surface

When Security Tools Become the Attack Surface

12 min read
23
Aug
Part 4 — The AI Supply Chain Can Execute Before the Application Does

Part 4 — The AI Supply Chain Can Execute Before the Application Does

AI artifacts are no longer passive data. DEF CON 34 shows how models, Skills, repositories and loaders can become execution paths — forcing CISOs to rethink supply-chain security, provenance and runtime authority.
15 min read
19
Aug
MCP Is Becoming Enterprise Infrastructure

MCP Is Becoming Enterprise Infrastructure

MCP is becoming enterprise infrastructure. As AI agents gain access to tools, identities and business systems, CISOs must govern not just integration — but the authority flowing through it.
15 min read
18
Aug
Part 2 —The Agent Is the New Attack Path

Part 2 —The Agent Is the New Attack Path

A safe model does not imply a safe agent. Agentic AI shifts the security problem from jailbreaks to authority: can an attacker make the system perform an authorized action for an unauthorized reason?
17 min read
08
Aug
The Perfect Blueprint: Why Digital Twins Have Become a Goldmine for Attackers

The Perfect Blueprint: Why Digital Twins Have Become a Goldmine for Attackers

Digital twins are transforming how buildings are designed and operated—but they are also creating an unprecedented reconnaissance asset. A compromised digital twin can reveal layouts, security controls and operational dependencies long before an attacker reaches the building.
5 min read