How to distinguish between "fictional consciousness" and "real consciousness"? Cognitive scientist Joscha Bach threw out this seemingly simple yet almost unanswerable question in a discussion, "Perhaps all consciousness is fictional - as long as you are not aware of it, it's a paradox." He admitted that this question is extremely complex on a phenomenological level and remains unresolved. "Fake phenomena of consciousness and real phenomenal experience are essentially virtual constructs."
Stephen Hsu
Stephen Hsu
Professor of Physics, Michigan State University
American physicist and entrepreneur. He received his bachelor's degree in physics from Caltech and his PhD from the University of California, Berkeley. Later, he taught at Yale University and the University of Oregon, where he served as director of the Institute for Theoretical Science, focusing on research in particle physics and cosmology. In 2012, Hsu was appointed Vice President for Research and Graduate Education at Michigan State University. However, his tenure was challenged by controversial views on genetic enhancement and racial classification, and he ultimately resigned from administrative duties in 2020. Additionally, Steve Hsu is the founder of Safeweb and Genomic Prediction, the latter specializing in developing genetic testing technology for IVF embryos. His work in genetics has made significant contributions to the genetic prediction of complex human traits and diseases.
Joscha Bach
Joscha Bach
PhD in Philosophy, University of Osnabrück
Joscha Bach is a German cognitive scientist, AI researcher, and philosopher, known for his work on cognitive architectures, artificial intelligence, mental representation, emotion, social modeling, multi-agent systems, and the philosophy of mind. His research aims to bridge cognitive science and artificial intelligence by investigating how human intelligence and consciousness can be computationally modeled.
Quick Takeaways:
1.The entire history of computer development has been chasing a "master algorithm": that mysterious entity capable of self-evolution and omniscience.
2. The reason brain cell populations can act as a unified subject is by constructing a "single-agent" control model, where cells extract environmental meaning and simulate a viewpoint: a virtual narrative about "how this subject should perceive the world."
3. I increasingly feel that the world is "code." The feeling of "directly experiencing reality," rather than us observing reality, is more like the brain generating models.
4. Human mental operation is essentially organized within multi-dimensional embedding spaces, where each dimension represents a modifiable parameter variable.
5. The essence of Earth life is not humanity, but the evolution of consciousness and life. I believe that without AGI, humanity is doomed to extinction.
6. Integrated Information Theory created so-called "axiom" chapters, which are actually just verbal descriptions of the definition of consciousness, not axioms in a mathematical sense, and the φ value is more like a theory promotion tool.
Growing up in the East German Forest
Stephen Hsu: We met at an event in Frankfurt in the summer. I had the chance to listen to several deep discussions you had about AI and also learned about your unique childhood experiences and growing-up story. So I'd like to start by talking about your early life, see how you got to where you are today, and then discuss the core topic of AI. Could you start by talking about your childhood spent in the forest?
Joscha Bach: My father was out of step with society, so he decided to create his own world. He was one of the first generation to do "social experiments," but he didn't recruit people to join, nor did he know how to expand his circle, and he wasn't even interested in these things. He was more like someone who built a private kingdom in the forest, only convincing my mother to support him, and they lived a kind of "Eden" life he created himself.
He was originally an architect, later became an artist, and believed the meaning of life was to create art and converse with his inner voice. Therefore, my upbringing was just a "byproduct" of his life; he didn't care much about the inner world of children. The place where I grew up was astonishingly beautiful, but I was also extremely lonely. Because I was bored, I started reading very early and absorbed knowledge like a sponge.
By the time I was supposed to go to school, I found myself disconnected from the outside world. At that time, East Germany practiced simple Marxist dialectics, and the content taught by the teachers was either something I had already read or found shallow and uninteresting. This experience led me to develop a kind of "inherent arrogance of usefulness." For a long time, I felt that others couldn't teach me much and that I had to learn everything myself. This didn't make me smarter, but it made me more unique, and it also made me accustomed to reconstructing the world using first principles to understand how things worked around me. I also found that others had different growth trajectories and psychology than me due to the process of socialization, meaning we had different "world interfaces." To communicate with them, I had to learn to "translate."
But unlike my father, I was very clear that living in seclusion in the mountains was not the destination I desired. Although it was quiet there, without the noise and excessive sensory stimulation of the city, I needed stimulation and interaction from others to push me to focus on my own projects.
Stephen Hsu: You now live in the Bay Area. Compared to the forest, do you think cities like San Francisco are more suitable for you?
Joscha Bach: I currently live in the South Bay. Although the neighbors are great and the weather is very good, it somehow combines the worst parts of both environments; it's neither remote enough to blend into nature nor urban enough to meet interesting people within walking distance. So I often drive to San Francisco.
Stephen Hsu: But the advantage is that many people who want to research or develop AI gather here.
Joscha Bach: Yes, that's exactly why I stay here.
Stephen Hsu: It seems your teachers back then couldn't spark your interest, but I'm curious if any historical thinkers or writers you read influenced you during your growth?
Joscha Bach: I read a lot of philosophy and science fiction. When I was young, I rarely read for entertainment; reading itself was interesting, but my main goal was to understand the knowledge system of society. At that time, I forced myself to read through the Bible and even Gandhi's writings, while also pushing myself to read math books and biographies of Einstein, etc. I read these just to understand the underlying logic of society. I thought everyone should master these classics - not just books, but also media like movies. I originally thought the essence of education was to make people deeply understand the society they were in. Later, I found out that almost no one does this.
▷ Source: Nix Ren
Stephen Hsu: Although I think your intelligence and personality are very unique, you did go through the standard education system and even got a PhD, staying in school for a long time, right?
Joscha Bach: That's right. I chose the academic path originally because I wanted to understand how the mind works. I examined various disciplines and found that none of them could answer this question alone.
At first, I was drawn to computer science because it fit my way of thinking very well. As a child, I was obsessed with programming, and I was lucky enough to be among the first generation of children to have a home computer. When I first got a Commodore 64, there was no ready-made software available on the market, so the first thing I had to do was write a text editor myself. Then I learned to create computer graphics from scratch. Looking back now, this experience shaped my way of thinking about building reality using computational frameworks.
After entering academia, I found that the way computer science communicated was very much to my liking. You could criticize a professor's proof in the first semester, and if you found a problem in their proof, everyone, including the professor himself, would be very grateful. But the philosophy department was completely different; the standards for acceptance there were highly socialized, and you had to fit into the circle. At least that was the case at the university I attended. Logic and analytical philosophy were slightly better, but the most hardcore analytical philosophy actually came from the mathematical logic courses in the computer science department. In addition, I also minored in related subjects like psychology.
While studying in Berlin, I would sometimes attend several universities at once. If I saw an interesting class - even a cognitive science class by a visiting professor at the University of Potsdam - I would drive to audit it. If the major courses weren't hardcore enough, I would just skip them, teach myself from textbooks, and only take the exams.
This highly autonomous learning benefited me greatly, but the