PERSPECTIVES

Thought leadership from industry experts

A hand interacts with AI agents on a laptop screen displaying various DEV OPS interfaces.

The highest risk for AI agents is the software development organization

Dennis ZimmerOctober 12, 2026

AI coding agents inherit broad developer privileges, demanding closer monitoring of autonomous actions.

Top Perspectives

Featured Columnists

Mike Britton
Abnormal Security, Chief Information Officer
Toby Gouker
First Health Advisory, Chief Security Officer
Callie Guenther
Critical Start, Senior Manager, Cyber Threat Research
Dustin Sachs
CyberRisk Collaborative, Chief Technologist and Senior Director of Programs
Stu Sjouwerman
KnowBe4, CEO
Morgan Wright
SentinelOne, Chief Security Advisor

More Perspectives

Futuristic Hand Reaching for AI Technology Interface Representing Artificial Intelligence and Data Management in a Digital World. Glyphic.

Don’t think of readiness as just an onboarding problem

Nick Heddy October 7, 2026

Real readiness happens when people, processes, and AI technology work in tandem.

Government security

A personnel file is a map: What federal personnel data breaches reveal

Cory SimpsonOctober 6, 2026

The government needs to protect personnel records as strongly as it protects its own secrets.

Cybercriminal with gloves creating fake website phishing scam with malicious web links and online fraud tactics. Cybersecurity threats, identity theft, data breach risks, and internet security danger.

MCP fixed secret handling. URL elicitation moved the phishing risk into the browser

Sunil GentyalaOctober 6, 2026

MCP URL elicitation protects secrets but shifts phishing risk into trusted browser flows.

(Adobe Stock)

AI benefits/risks

Why it will take strong public-private partnerships to fight AI-based threats

Derek MankyOctober 6, 2026

The AI threats are coming at us too fast for any one company or country to go it alone.

Futuristic digital security concept with robotic hand reaching for file icon and login credentials over network connections in a modern city backdrop. Sentient.

NVIDIA can stop a rogue agent. Can it revoke its access?

Garrett GrossOctober 5, 2026

Quarantining rogue AI agents isn't enough if their credentials and access remain active.

LLM technology integrated into complex circuit board with illuminated traces and central processing unit

How identity and permissions become the blast-radius boundary for LLMs

Art PoghosyanOctober 5, 2026

Stop trying to secure every last LLM – here’s how limiting permissions become the last line of defense.

Futuristic Hand Reaching for AI Technology Interface Representing Artificial Intelligence and Data Management in a Digital World. Glyphic.

AI benefits/risks

Four ways to close the AI security gap

Mick Leach October 2, 2026

When it comes to deploying AI, take a page from the military, where I learned: “slow is smooth, and smooth is fast.”

For Cybersecurity Awareness Month, it’s time to rethink zero-trust for the AI era

Danny JenkinsOctober 1, 2026

Here’s three ways to apply zero-trust principles to AI deployments.

Zero Trust - Zero Trust Network - Zero Trust Architecture - ZTA

Zero Trust ends at the browser

Hadar Blutrich October 1, 2026

Zero Trust often stops at the browser, leaving third-party scripts and sensitive data exposed.

patch presented in the form of binary code

Patch/Configuration Management

Stop patching by score – and start patching by authority

Kevin E. GreeneSeptember 30, 2026

When we understand the authority an exploit gives a threat actor and what it lets them do next, we understand why “patch now” makes sense.

Upcoming Webcasts