OpenAI's o3 model aced a test of AI reasoning – but it's still not AGI

The latest AI model from OpenAI achieved an “impressive leap in performance” but it still hasn’t demonstrated what experts classify as human-level intelligence