A few days ago, I woke up to the news that Slavoj Žižek wrote an article published on Die Zeit (behind paywall) about my latest work, the Infinite Conversation. I was elated to realize that the project had eventually landed on his desk, and overjoyed to see that he opened his article, in perfect Žižekian style, with a fecal reference.

There’s a lot in his piece that I agreed with: his acknowledgment that much of postmodern philosophy reads as muddled, facilitating the task for an AI attempting to “write” about ideologies; the fact that lack of clarity can sometimes be interpreted as profound ambiguity, an effect exploited by the Infinite Conversation; the realization that the voices of Herzog and Žižek can serve as background soundscape, as indeed many people report using the Infinite Conversation to peacefully fall asleep.

I could even agree with the notion that the Infinite Conversation will not come up with a real theory. The key here is the “come up” part, since AI language models do not have intentionality. They can only react to what we ask from them. But they can already do so at staggering speed and that is sometimes a useful task in it of itself. It’s a bit like the infinite monkey theorem, except that the digital monkeys already know all of Shakespeare by heart and can continuously produce sonnets in his style that often make sense. A poet could sift through dozens of such generated sonnets and pick one that speaks to them, because Meaning is always in the ears of the beholder.

For instance, I asked AI-Žižek what he thought about being parroted by a chat-bot using his voice, and this is the best cherry-picked answer out of 3 generated replies:

The first time I heard that my voice was imitated by some artificial intelligence, I was deeply hurt. But then, when I learned more about it, my reaction was the opposite. I think that in the future, this will happen to all of us. The idea that you are yourself, that you are a unique individual with your own personality, will disappear much earlier than we usually think. Already today there are algorithms that can predict what you’ll answer to certain questions. You know the type of questions: What’s your type, do you like this or that, do you want to go on a date, and so on. I think that in a couple of decades we will enter a period when you will not be able to distinguish between a real person and an artificial intelligence. And I think that this will be a big shock for people. Here, I think, we have an extremely interesting paradoxical situation. The idea that we are unique individuals with a unique personality is disappearing. We are part of big networks where we are interchangeable.

The only part that I truly found dissatisfactory about the article is the conclusion that my work does not present a suitable opportunity to raise awareness and reflect on the dangers of this technology. In fact, his response is in a way a fulfillment of that goal, in that he has continued the discourse and contributed in spreading the word. I would be curious to hear him expand on his reasoning, and can provide an opportunity for him to do so. I invite him to take part in a real, live conversation to be held at the Misalignment Museum, a space currently being set up in San Francisco that will display art, including the Infinite Conversation, as well as host events and talks around the cultural aspects of AI. Now if only we could convince Werner Herzog to join too!