Human-level general AI is a long way off, and today's fear of AI causing human extinction is overblown. AI is a powerful general-purpose tool — like electricity — that we should build and deploy as widely as possible.
Progress is real but genuine AGI is likely many decades away. Fear-driven, heavy-handed regulation would slow enormous benefits in health, education and productivity while doing little about the speculative risks.
Concrete evidence of imminent, autonomous, broadly superhuman systems — not extrapolation from today's narrow models.