The Fable 5 Shutdown: What a 19-Day Export Control Taught Us About AI Risk Governance
Back in April, I wrote about Project Glasswing — Anthropic's decision to release its first Mythos-class model only to a vetted group of cyber defenders and critical infrastructure operators. At the time, I called it an interesting experiment in tiered access control for frontier AI. I didn't expect the sequel to involve the U.S. Commerce Department, an Amazon-discovered jailbreak, and the fastest deploy-suspend-redeploy cycle in frontier AI history. But here we are. Claude Fa
Jul 1
Anthropic’s Claude Fable 5 and Mythos 5: When AI Becomes Too Powerful to Release
Introduction For the first time, an AI vendor has publicly released a model while simultaneously admitting that its unrestricted version is too powerful for general availability. On June 9, 2026, Anthropic announced two new frontier AI models: Claude Fable 5 and Claude Mythos 5. While both models share the same underlying architecture, Anthropic made a deliberate decision to split them into separate offerings. Claude Fable 5 is available to the public with extensive safeguard
Jun 10
The 2026 Verizon DBIR Closed the Window. Your AI Endpoints Are Still Open.
The cybersecurity industry has long relied on the Verizon Data Breach Investigations Report (DBIR) to understand how attackers gain access to organizations. The 2026 report reveals a significant shift that security leaders cannot afford to ignore. For years, stolen credentials and phishing dominated the conversation. Today, vulnerability exploitation has become the leading initial access vector in confirmed breaches. This change represents more than a new statistic—it highlig
Jun 9
