OSINT stands for open-source intelligence. It is the collection, analysis, and dissemination of information from publicly available sources, such as social media, government reports, newspapers, and other public documents. OSINT is commonly used by intelligence agencies, private investigators, and law enforcement to gather information about an individual or organization. The OSINT framework showcases the multiple ways in which organizations can gather intelligence.

Security teams are under more pressure than ever, reacting at human speed while systems, identities, and AI agents operate at machine speed. In this episode of Security Matters, host David Puner sits down with cybersecurity leader and former FBI executive MK Palmore to explore why defenders struggle to keep pace and what it takes to regain control.

On March 03, 2026, pac4j released fixes for a maximum severity vulnerability in pac4j-jwt, tracked as CVE-2026-29000. The flaw arises from improper verification of cryptographic signatures in the JwtAuthenticator component when processing encrypted JWTs (JWE). A remote, unauthenticated threat actor who knows the server’s RSA public key can bypass authentication and impersonate arbitrary users (including administrators) by submitting a crafted JWE whose inner token is an unsigned PlainJWT.

BlueVoyant Security Operations Center (SOC) and Threat Fusion Cell (TFC) continue to track an activity cluster that uses email bombing and IT-support impersonation over Microsoft Teams to obtain Quick Assist access, then pivot to a deeper attack. This research shows that once on the victim’s host, the actors sideload a malicious DLL to deliver a new backdoor BlueVoyant has dubbed the A0Backdoor.

In today’s smart factories, production doesn’t go quiet at shift change. Behind the scenes, modern manufacturing systems never cease. They continuously exchange data, adjust software and processes in real time, and allow vendors to connect remotely to monitor performance or deliver updates. As these interactions multiply, the number of identity-driven points grows just as quickly.

Most data breaches don’t happen because systems fail. They happen because people make routine errors. Attackers know this, which is why social engineering has become the dominant attack vector, exploiting everyday actions like emailing files or responding to messages. Today, 70–90% of successful cyber attacks involve social engineering, resulting in data exposure that technical safeguards can’t intercept.

If you’re building AI applications that touch sensitive data, tokenization isn’t optional. It’s the layer that decides whether your pipeline leaks PHI, PII, or financial data to your LLM, or keeps it protected. But here’s where most teams stop thinking: not all tokenization is the same. Two approaches you’ll encounter most often are entropy-based tokenization and polymorphic tokenization. They sound similar. They serve completely different purposes.