Could A.I. Write a Novel Like Hemingway? | Salman Rushdie

You know, I mean, I never say never, you know, but I mean, I remember, I mean, I've sort of an amateur chess player. It's what I'm interested in—chess. And I remember back in the day when computers were first being taught to play chess, that people would say that they would never be able to beat the real great— that the grandmasters, you know, and the world champions.

And for a long time, that was true; the world champion players, the great grandmasters, were able to, you know, to overcome the computer. Not true anymore. No, it's not true anymore. The computers are certainly as good, if not better, than any human player. As computer memory and sophistication has increased, you know, it's outstripped human memory and sophistication.

So, I don't know everything. It seems to me the thing that makes a writer a good writer, you know, is not just the technical skill with language, not even being able to find and tell a good story. You know, it seems to me that, first of all, there's a relationship with language that the best writers have, which is very much their relationship. You know, if we read Hemingway, we know it's Hemingway—that's it—because he has a particular relationship with the language.

And if we read James Joyce or William Faulkner, we know it's them. And if we read Garcia Marquez, you know, same thing. So, that's the first thing. When I'm looking at work, I've tried to see what is the relationship with language.

And the second—the second thing are, you know, how you see the world. But do you have a good ear? I mean, are you good at listening to how people really speak? Do you have a good eye? Are you good at seeing the world in an interesting way? You know, and then finally, the greatest writers, the best writers, have a vision of the world that is personal to themselves.

They have a kind of take on reality, you know, which is theirs, and out of which their whole sensibility proceeds. You know, now, to have all of that in the form of artificial intelligence, I think it's quite—it's quite—I don't think we're anywhere near that yet. But what is true, I think, is that there's beginning to be some sense of AI as developing and a moral sense developing—an ethical ability to make good and bad, good and evil choices, right and wrong choices, you know.

And that's a step on the way towards being what one would call human. So, I mean, I'm not saying never; I'm just saying I don't see that we're there yet.