Stop Calling Everything AI, Machine-Learning Pioneer Says

While I mostly agree with Michael Jordan, I think that he overstates his argument a small bit. I feel comfortable applying the "AI" label to projects if they: 1) achieve human level of performance or better. 2) They use any or all of standard techniques like deep learning, NLP, knowledge representation, reinforcement learning, etc.

I am just about to hit my 40 year anniversary for getting paid to be doing "AI" related work, with a lenient definition of AI.

On the other hand, Artificial General Intelligence is a much stronger term, and is something that I don't expect to see in my lifetime.

AI isn't really a thing per se and it adds a kind of patina of false mystique to computation that is essentially no different than any other computation (which itself is also not really "a thing" out there in the world as a phenomenon). The only way AI really means anything is because the person in question has decided to frame something as AI. He is choosing to append to something meaning that it itself does not possess. Take a the linear regression type stuff lots of AI today uses. You put that in another context and it ceases to be AI. Why? Because it's not AI per se! If it were a real thing or real phenomenon, it would be mind-independent and the mental context wouldn't determine its identity.

Basically, it seems that whenever someone mechanizes something that previous only a thinking person could do, then its viewed as AI.

The rebuttal to this view is "provide a principled definition of intelligence". Doesn't seem like the article does this.

A hint appears partway through: "computers will not be able to match humans in their ability to reason abstractly about real-world situations". Does human intelligence distinguish itself by its "abstractness" and by its application to "the real world"? There's also "the systems do not form the kinds of semantic representations and inferences that humans are capable of". Seems like a promising direction for some definition of intelligence.

For my money, we will never consider any machine intelligent as long as we mass produce it. The only way we'll accept machines as intelligent is if, as the singularity theorists say, the machines build themselves. Then we aren't really mass producing them, we're just kicking off a process that we don't totally understand, a bit like gestation.

Products are shipped which claim to use "AI". This is self-contradictory with this definition, and no product can ever exist claiming AI with it. I'm fine with that, but then it seems that the term would be completely useless, when "science-fiction" already exist (and, even if it is not always, can be realistic)

"Ok, we'll call it Quantum instead." /s