Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

AI agents and identity risks: How security will shift in 2026

The pace of technological change is relentless. Not long ago, our migration to the cloud and the automation of CI/CD pipelines dominated the conversation. Now, AI agents are reshaping how we think about automation, productivity, and risk. As we look toward 2026, it’s clear that these intelligent, autonomous systems are not just a passing trend; they are becoming foundational to how businesses operate.

TLS certificate management in 2026: The endless game of Whack-A-Cert

As 2025 races to a close, you’ll see several predictions about AI agents, quantum computing, and other frontier innovations. Don’t get me wrong, I’m excited about solving these challenges, too. But there’s a quieter, less flashy countdown underway, one that will determine whether organizations can even reach the cutting edge. TLS certificates—the machine identities used to prove machines are who they say they are—will begin expiring twice as fast in March 2026.

EP 20 - Why agentic AI is changing the security risk equation

As enterprises embrace agentic AI, a new security risk equation emerges. In this episode of Security Matters, host David Puner sits down with Lavi Lazarovitz, VP of Cyber Research at CyberArk Labs, to unpack how AI agents and identity security are reshaping the threat landscape. Learn why privileged access is now the fault line of enterprise security, how attackers exploit overprivileged AI agents, and what security teams must rethink before scaling AI.

When cybercrime meets cyberwarfare

Across today’s threat landscape, the divide between cybercrime and cyberwarfare is disappearing. Financially motivated groups and state-sponsored actors rely on the same tactics, techniques, and procedures (TTPs)—exploiting zero-day and one-day vulnerabilities, abusing ransomware-as-a-service (RaaS) platforms, hiding behind proxies, and living off the land (LotL) within legitimate IT environments. They also often target the same enterprises.

The next chapter of identity security begins with privilege

Privileged access management (PAM) was once thought of in simple terms: secure the credentials of a handful of administrators managing on-premises systems. Vault the passwords, rotate them regularly, and record every privileged session It worked for a world with clear boundaries and predictable users. That world is now a museum piece. But here’s the shift: It’s not that PAM has changed. The very definition of privilege has evolved.

Racing and Fuzzing HTTP/3: Open-sourcing QuicDraw(H3)

This blog post provides a dive into HTTP/3’s evolution for security engineers, an overview of our research journey, and what led us to develop the open-source tool QuicDraw, which can be used for fuzzing and racing HTTP/3 applications. QuicDraw implements “Quic-Fin-Sync” our implementation of the last-byte-sync with the single packet attack on HTTP/3. We conclude by evaluating QuicDraw’s performance against a real-world target and comparing its results to other tools.

Why access management needs a challenger mindset

Cybersecurity never stands still. Every login, session, and connection shifts the balance between freedom and control. Effective access management today isn’t about restriction—it’s about enabling trust at the speed of innovation. Modern enterprises achieve this by evolving their controls to be seamless, adaptive, and invisible to the user.

How AI is reshaping identity governance for CISOs and CIOs

2025 has been a defining year for identity security, marked by a rapid increase in the volume, variety, and velocity of identities that organizations must now govern. The changes have been building for a long time, as identity tools have evolved from early single sign-on solutions and compliance-driven governance to the cloud-first, AI-powered world we live in now, which must enable employees with the access they need at lightning speed while maintaining security.

EP 19 - Trust under attack: Spies, lies, and the new face of cybercrime

Eric O’Neill, former FBI ghost and author of “Spies, Lies & Cybercrime,” joins host David Puner to take a deep dive into the mindset and tactics needed to defend against today’s sophisticated cyber threats. Drawing on O’Neill’s experience catching spies and investigating cybercriminals, the conversation explains how thinking like an attacker can help organizations and individuals stay ahead.