I want to share my experience in the cloud, AI, and agents development areas.
What if companies without human employees could actually be good for people? This thought experiment questions how we treat employees, whether some entrepreneurs should only manage AI agents, and what responsibility companies still have towards society.
What would an AI need to play music with us in real time? As a drummer and AI engineer, I followed that question from ChatGPT singing to multimodal AI and humanoid robots.
Trustworthy Agentic AI is not about eliminating risk. It is about knowing when a system should act, ask for more information, or abstain. In this follow-up, I connect my earlier risk-taking argument with Jev, bounded probabilistic decisions, confidence thresholds, blast radius, deterministic policy logic, and the question of whether every decision inside an AI agent really needs a generative LLM.
