Explore recursive self-improvement, agentic AI risks and the AI singularity debate, including OpenAI security incidents and ...