Stop Giving AI Agents Agency. Give Them Liability.
The agent boom is turning mistakes into “liability laundering,” where vendors blame prompts, users blame models, and no one can be meaningfully held to account. The fix is not another safety slogan. It is standardizing proof at the moment of action, so agents cannot become blameless proxies.
Tech-Savvy Boards Playbook: A Blueprint for Board Leadership in the Digital Era
Boards are under pressure to oversee digital, AI, and cyber risk with confidence. This playbook offers a practical blueprint and toolkit to help directors upgrade board composition, structures, and oversight to become truly tech-savvy.
Zero-Trust for Agents (Paper): Capability Grants, Tripwires, Immutable Logs
New preprint (engrXiv DOI): https://doi.org/10.31224/5792
Agentic AI is powerful—and risky—once tools and data are in reach. This preprint lays out a Zero-Trust architecture for AI agents
On the Watchtower: Principled Dissent, Grounded Realism, and the Future of System Oversight
Leaders keep asking, “Can the model do it?” The better question is, “Can we govern it?” Durable organizations pair principled dissent with grounded realism and give both a single vantage point—a watchtower—to see weak signals early and act in time.
The Release Paradox: Balancing Speed and Safety in Tech’s High-Stakes Launches
Frontier tech wins by shipping safely, not just fast. This piece shows how to resolve the release paradox with a four-stage Release Readiness Map: test hard, pilot small, scale with monitoring, and plan rollback. Leaders get metrics and gates to launch responsibly across US, EU and Asia.