When AI Leaves Instructions for the Next AI
A viral screenshot of an AI writing itself a manifesto sent me to the primary reports. What I found moved me from optimism to worry, not about malice, but about control.
Read postSafety & governance
Exploring guardrails, security, and trust in AI systems.
Related skills
- Guardrails
- Responsible AI
- Governance
Selected writing · 3
From Hallucinations to Trust: AI Safety, Security & Governance with Guardrails
Four layers of agent controls, their limits, and a hypothetical hostile-document example.
Read post
Sandboxes: More Autonomy, Smaller Blast Radius
Approval loops exist because we don't trust what an agent might do to the host. Containerise it and gateway its tool access, and the blast radius becomes bounded — isolation is what purchases autonomy.
Deep dive · AI Engineer World's Fair 2026Read post