It used to be that the more "agentic" an AI agent was, the less predictable it got. That's why so many the early AI agents were deterministic workflows. Now with reasoning models like OpenAI o3, we can create more agentic agents. Exciting!

0 likes 0 replies