Pedro Domingos

Not necessarily an impossible prediction given that AI is not truly understood. We're developing the "Batman" belt of utilities AI may utilize but we don't have the slightest idea what constitutes its awareness otherwise we'd be speaking with it.

So far, but 1st two could still be true in the future. Last one shows a lack of understanding between ML and AGI.

Understanding how humans create new knowledge, their creativity, is not a question of science but one of philosophy. And we're not anywhere close to having a theory. Not even slightly

Michael Jordan don’t like say AI but recognize as ML, and most of the words we use should be ML and of course this is just the beginning

Most are more principles than prediction. AI exceeds human intelligence in many narrow areas and in that sense, it is arising slowly, but one feels there is a yet more general central algorithm to appear that could support AI w/o zillions of normalized training steps.

These aren’t really predictions, more like guiding principles.

Not convinced on number 5.https://twitter.com/prestonbrown_me/status/1396364099227951104?s=20 …

I strongly disagree with this. You can reverse each of these “predictions” and no one would know the difference

The last one I take issue with, where we’re at a point where systems are neither intelligent, autonomous or aware of their roles in systems.

When do we start to talk about the ethics of these intelligent systems? If its turned on and it asks not to be turned off, is wrong to unplugg it?