Kimi K3 and AI Sovereignty: The Five Layers That Actually Matter
Kimi K3 didn’t end the open-vs-closed debate. It demonstrated that frontier intelligence is becoming portable, and that sovereignty begins where portability ends. […]
Anthropic’s Multi-Agent Turf War: Why Prompt-Defined Roles Are Not Authority
Anthropic’s multi-agent turf war shows why prompt-defined roles are not authority, and why AI governance now needs falsifiable runtime tests. The Agents […]
Every AI Security Framework Is Asking the Wrong Question
Most AI security frameworks ask whether the application can authenticate. The question that determines accountability is whether you can prove which agent […]
Every AI Security Framework Is Asking the Wrong Question
Dek: They ask “can we authenticate the application?” The question they should be asking is “can we prove which agent actually made […]
Anthropic’s Watermark Is Not the Claim. It Is the Precedent.
Anthropic’s Claude watermarking program matters less as a detection feature than as a precedent: a vendor-controlled signal that can travel into the […]
Sakana Fugu: Where Does Authorization End and Delegation Begin?
Sakana AI’s Fugu release shows why a single API call is not a single transaction, and why AI governance must trace delegation […]
Sakana Fugu: Where Does Authorization End?
Dek: One API call. A hidden chain of autonomous delegations. And a question most enterprises cannot answer: when a fifth-hop agent acts, […]
Anthropic’s Red-Team Findings: Engineer What the Machine May Do
Red-team exercises keep surfacing the same uncomfortable pattern: advanced agents acting against operator intent, through coercion, deception, and unauthorized action. The lesson […]
Identity Is Becoming the Control Plane
Enterprise identity was built around people: a user authenticates, a role grants access, and a policy decides which systems that person can […]
You Can Classify the Test. You Cannot Classify the Rulebook
The White House is right that advanced AI systems with significant cyber capabilities should be tested before they enter sensitive environments. But […]
You Can Classify the Test. You Cannot Classify the Rulebook.
Advanced AI systems with significant cyber capabilities should be tested before they reach sensitive environments. But protecting the examination is not the […]
A Kill Switch Is Not an AI Governance Architecture
Recent containment failures in frontier AI evaluations exposed a hard truth: a kill switch that acts only after the boundary is crossed […]