From OpenAI, commented on how current LLM limitations do not necessarily preclude achieving AGI.
AI Explained