This week Anthropic announced Project Glasswing, a cybersecurity initiative built around Claude Mythos Preview, an unreleased frontier AI model capable of autonomously discovering and developing exploits for zero-day vulnerabilities across major operating systems and web browsers. According to early details, the model has already identified thousands of critical vulnerabilities that traditional tools have missed for years.

Ask anyone what they think when a website requests a driver's license, Social Security number, or email address, and you'll hear the same reaction: "Why do they need that?" It’s a fair question. Not a day goes by without news of another data breach or scam. Many people have either experienced fraud firsthand or know someone who has. While they're more aware of the need to protect their data, they don't feel equipped to actually do it.

AI is making phishing attacks easier to create and scale. Tasks that once required manual effort can now be automated, allowing attackers to generate realistic messages, launch campaigns, and adapt tactics quickly to evade security controls. In fact, KnowBe4’s 2025 Phishing Threat Trends Report found that more than 73% of phishing emails analyzed in 2024 showed signs of AI involvement. As a result, phishing threats are becoming harder to detect using traditional methods alone.

The software supply chain continues to face escalating threats, with malicious actors targeting developers and organizations at an unprecedented scale. In our Spring 2026 Threat Research Review, we analyze the latest trends, uncover alarming statistics, and highlight the evolving tactics used by attackers. From dependency injection attacks to the rise of typosquatting, this report provides a comprehensive look at the threats shaping the software ecosystem.

My first introduction to UNIX remote access was via telnet and rsh protocols in college, which was the standard method at the time. But I soon started reading articles about how easy it was for someone to sniff the network and capture passwords since they were being transmitted in plaintext. On the shared network segments common to university campuses and early enterprise environments, the tools to intercept traffic were freely available, well-documented, and required very little skill to use.

We can't control the pace of AI-driven vulnerability discovery, but we can control how fast we respond. Last week, Thomas Ptacek published a piece arguing that vulnerability research is cooked. His thesis: AI agents are about to drown us in a steady stream of validated, exploitable, high-severity vulnerabilities, faster than anyone can patch them. But from where I sit, the more urgent question isn't whether the flood is coming, but whether the infrastructure we depend on can absorb it.