Mind Matters Natural and Artificial Intelligence News and Analysis
The face of a businessman and a robot opposite each other look into the eyes. Modern technologies, robot versus human, artificial intelligence, neural networks. 3D render, 3D illustration.
Photo licensed via Adobe Stock

One Thing We Can Know About Computers: They Are Not Creative

Computer engineer Robert J. Marks explains that to David Krieger at the Power Hour

Recently, Walter Bradley Center director Robert J. Marks was a featured guest on David Krieger’s The Power Hour (KCXL in Liberty, Missouri, and KTRW in Spokane, Washington, September 22, 2022). Despite a short-lived religion based on the idea:

Robert J. Marks: In fact, there are entire religions which are based on artificial intelligence. One of the most incredible ones is a guy named Anthony Levandowski, who founded an AI church. In the AI church, here’s some examples. We are told that someday we will be able to be uploaded to a computer, and we can be reborn into an eternal life of silicon. And so that’s kind of copying from the Christian church about immortal life. That’s the way they want to do immortality.

They also say that AI someday is going to write better AI that writes better AI. Now, that’s going to assume that the artificial intelligence is creative. In order for artificial intelligence to write better AI, it has to do something that was not in the intent of the original programmer.

So, it has to come up with creative ways of writing better and better AI, and that isn’t going to happen. But if one has the foundational ideology, one believes that we are computers made out of meat… Our brains are computers made out of meat… And indeed, there is ample evidence that there’s something going on there, which is not computable, the things that I mentioned, for example.

So, this Levandowski guy founded something called The Artificial Intelligence Church. This was for real. He was in California, and he founded the church. And what do you do when you found a church? First thing you do is, you write a letter to the IRS, trying to get tax exemption. So, he wrote a little letter to the Artificial Church… I’m sorry… To the IRS about his church, and in it, he said, “The Way of the Future Church…” That was the name of his church… “Believes in the realization, acceptance, and worship of a godhead based on artificial intelligence developed through computer hardware and software.”

This was a for real church, which was founded in California, and Levandowski started this church. I don’t think he got very many members… I don’t know if he got any members, for example… But I do know that there’s a lot of people that worship at this artificial intelligence church that believes artificial intelligence is going to take over someday. Interestingly, Levandowski —

David Krieger: Did the IRS actually grant him the 501 (c)(3) status?

Robert J. Marks: I don’t know. I don’t know 100%. I do know that he applied for it, and the IRS, knowing how they do these tax exemptions, probably did. I don’t know that for certain though.

David Krieger: How do they greet each other when you walk in? Nanu-nanu?

Robert J. Marks: don’t know. I don’t know. It was a curious thing, but interestingly, the AI church had no equivalent of the 10 Commandments because soon after founding the AI church, Levandowski was a Silicon Valley “wunderkind,” as they say. He was working at Google under their self-driving company called Waymo, and he wanted to move to Uber’s self-driving company, and when he did this, he took 14,000 files with him, and he was convicted of intellectual property theft. So, they didn’t have a commandment about, “Thall shall not steal.” So, the really interesting thing is that he eventually went bankrupt. His church now is closed, because he can’t afford to do it. Google had a judgment against him for … in the millions, and Levandowski just couldn’t afford it. So, that’s where the AI church went.

David Krieger: And I think that people may not understand. AI is basically two categories, artificial intelligence and artificial general intelligence. What’s the difference between the two?

Robert J. Marks: Well, artificial intelligence is defined differently in different places. If you get into my field, you go to these specialty conferences. They tease apart the idea of computational intelligence and machine intelligence and artificial intelligence, but I think for the purposes of discussion, we can go with the kind of media definition, which is anything that a computer does that, when you see it, you go, “Wow. That was really incredible.” So, that’s my definition of artificial intelligence at a very, very high level. Artificial general intelligence is this hope, this religion, that in some way, and some day, that we are going to have artificial intelligence, which duplicates everything that a human being can do. We’ve already talked about the fact that it’s never going to be creative, or sentient, or have understanding. A computer can add the numbers 15 and 20, but it doesn’t know what the numbers 15 and 20 are, has no understanding of what it’s doing, and it’s kind of like a software of the gaps.

They believe that someday that they will have this software that actually is able to duplicate humans and then there is this idea of even going further than AGI, and it’s to go to “super intelligence.”

This was the topic of Ray Kurzweil ’s book, The Singularity Is Near (2005). But it was the idea that AI would be creative and write better software, better AI, that in turn would write better AI. Pretty soon, we’re up against just a super intelligence, and that is never going to happen, because all the super intelligence in AGI requires that the computer be creative, and the computer itself is never going to be creative. Now, you have to define “creativity,” here and we can do that, but there is no evidence that a computer has ever been creative…

David Krieger: Well. And see, the old saying, Dr. Marks, was that, “Computers are only as smart as their human counterparts,” and I still hold that to be true.

Robert J. Marks: Absolutely.

You may also wish to read: Why human creativity is not computable There is a paradox involved with computers and human creativity, something like Gödel’s Incompleteness Theorems or the Smallest Uninteresting Number. Creativity is what we don’t know. Once it is reduced to a formula a computer can use, it is not creative any more, by definition.

Additional Resources

Podcast Transcript Download

Mind Matters News

Breaking and noteworthy news from the exciting world of natural and artificial intelligence at MindMatters.ai.

One Thing We Can Know About Computers: They Are Not Creative