My perspective is that the Turing Test is now obsolete. Two months ago Peter Norvig and a friend of his wrote an article arguing that the goal posts have moved

My perspective is that the Turing Test is now obsolete. Two months ago Peter Norvig and a friend of his wrote an article arguing that the goal posts have moved in measuring AI. I have been a paid AI practitioner since 1982, and when I watched my niece’s family last night passing around my iPhone running ChatGPT Pro, just the looks on their faces as they were having spoken conversations, in all matter of topics, and asking for images to be generated and refined, I felt that AI is here. LLMs remind me of Plato’s metaphor of humans only seeing the shadows on a cave wall, not true reality. LLMs are trained on text and images and are disembodied. The learn from an information lossy projection, but still an interesting “reality.” We are likely to get real AGI when APIs are embodied, and toss in a few new technologies and optimizations to run cheaply on edge devices.