AI Safety Cannot Be Voluntary
For most Americans, warnings about AI safety have long sounded abstract, but that is quickly changing. Months of reports about
AI Safety Cannot Be Voluntary Read Post »
Govern
A Different Lesson from the OpenAI Attack on Hugging Face: Start Governing Before Production
Governance obligations concentrate on deployment. The operational controls in NIST AI RMF, the management-system requirements in ISO 42001, and the
From Shutdown to Standard: The Birth of the Cyber Jailbreak Severity Framework
The most consequential AI governance artifact of 2026 was not produced by a standards body, a legislative committee, or a
From Shutdown to Standard: The Birth of the Cyber Jailbreak Severity Framework Read Post »
Holding Back the Model Was the Easy Part
A single company can pause their own model. Slowing an industry that rewards whoever refuses to stop takes enforceable rules,
Holding Back the Model Was the Easy Part Read Post »
Guardrails Are Behavior – Orchestration Is Control
Today the New York Times reported that researchers in Italy bypassed the safety controls on 31 large language models using poetry. Wrap
Guardrails Are Behavior – Orchestration Is Control Read Post »
Connecticut’s New AI Law: The Good, the Bad, and What Still Needs to Be Done
Connecticut lawmakers deserve credit. On May 1, 2026, the House voted 131–17 to pass Senate Bill 5, the Artificial Intelligence
Connecticut’s New AI Law: The Good, the Bad, and What Still Needs to Be Done Read Post »
Anthropic’s Responsible Scaling Update – A Necessary Adjustment, But a Dangerous Signal
I’ve viewed Anthropic as the most structurally serious of the leading AI companies when it comes to safety. Their original
Anthropic’s Responsible Scaling Update – A Necessary Adjustment, But a Dangerous Signal Read Post »
Guide to updating from NIST CSF 1.1 to 2.0
The newly released update in early 2024 of the NIST Cybersecurity Framework (CSF) from 1.1 to 2.0 represents a significant step forward
Guide to updating from NIST CSF 1.1 to 2.0 Read Post »