Post

FSB wants "more human oversight" for agentic AI. The same AI that escaped OpenAI's own controlled tests? They admit these agents don't reason like us. But somehow, *we'll* control them? Get real.