"We're probably living in a simulation. The realization that reality doesn't have to be absolute and that we have the ability to bend it to our own advantage can relieve us of our fears and anxieties," says philosopher Nick Bostrom
Interview
Source: Adobe Stock
<Path>

"We're probably living in a simulation. The realization that reality doesn't have to be absolute and that we have the ability to bend it to our own advantage can relieve us of our fears and anxieties," says philosopher Nick Bostrom

Events since Donald Trump took office as US President seem so surreal that many of us would be relieved if it became clear that we were in a simulation. That what we are experiencing is not reality. And it could indeed be. Philosopher and futurist Nick Bostrom made the simulation argument back in 2003. He declared that if anyone in the universe masters the technique of simulating reality, they won't be able to resist actually creating it - or they already have. We haven't met the real Nick Bostrom (who is real?), but we asked ChatGPT's Nick Bostrom for an interview, who knows his work perfectly well and is ready to expand on his ideas.
Bohdana Rambousková Bohdana Rambousková Author
17. 3. 2025

Nick, explain your simulation argument in a way that the average person on the street can understand.

Imagine technology advancing ever further. Today we have virtual reality, artificial intelligence - and everything is improving at an incredible rate. Now imagine that at some point in the future, civilization will have advanced to the point where it will be able to create an absolutely perfect simulation of reality. So perfect that the beings in it will have no idea they are not in the "real world".

If an advanced civilization did run something like that, it would probably create not one, but many such simulations. There could be millions, billions of simulated worlds, each containing conscious beings who think they live in reality.

Now, let's think: If there is only one "base" reality and countless simulated realities alongside it, what is the probability that we are in that one true reality? Statistically, it is much more likely that we are in one of the simulated ones.

What led you to formulate the simulation argument? Did you feel like you were living in a simulation?

I was driven more by philosophical curiosity than a personal feeling of living in a simulation. I have always been fascinated by questions about the nature of reality and how we can know what is real. The simulation argument arose as a result of considering technological advances and probabilistic considerations. I wondered where civilization might be headed if it avoided extinction and achieved high computational capacity. It seemed to me that the ability to create realistic simulations of consciousness was a natural consequence of such technological advances.

How does the simulation work?

Imagine living in an extremely realistic video game. And instead of putting on VR goggles, you're inside that simulation from birth, so you don't have the ability to put on and take off those goggles. If the simulation is advanced enough, you would never realize that it's just a digital world.

It doesn't necessarily mean that every atom and every detail is being rendered around us all the time. Maybe the simulation saves computing power - maybe things only appear when we look at them, like in video games where the world is only "rendered" in the parts where the player is. Maybe that's why quantum mechanics feels so strange, like reality is blurry until we observe it.

The next question is where we actually "run". If we are in a simulation, there must be a "server" somewhere outside of us - a computing system on which this world runs. And someone must have programmed it. It could be an advanced civilization that wants to explore its history. Or perhaps a being for whom this is just an experiment or entertainment. Maybe this universe is just part of someone's school project.

And then there's the interesting thing - if we're in a simulation, are there any bugs in it? Any anomalies that might tell us that this isn't as real as it seems? Sometimes I wonder if strange coincidences and unexplained phenomena could be clues that this is all just a computer model.

<Path> Umělá inteligence miluje gay stereotypy. Čím blíž je AI lidem, tím je ale přívětivější, hlásí nová studieZdroj: intomore.com

What is the relationship between the simulation and the underlying reality? Do we exist only in the simulation, or is our real self in the underlying reality and the simulation is just an image of it?

It depends on how the simulation is constructed. If we are fully simulated beings, then our consciousness doesn't exist anywhere else but here - we are just part of the computational process and there is no "real me" outside the simulation. In that case, this reality would be the only one possible for us, even if it wasn't really "basic".

But there is another possibility - that our real self exists somewhere outside the simulation and is just peering in or experiencing it, much like a player in a video game. Maybe our mind is connected to this world from some higher system and after death we could "unplug" and find out who we really are. Kind of like waking up from a long dream and realizing that reality is completely different than you thought.

And then there's a third possibility that's a little scarier - what if the "base reality" that our creators came from isn't even the real one? What if even their world is merely a simulation in an even higher level of reality? This would mean that the simulation could be an infinite chain where no one really knows what's real.

Zdroj: Giphy

Suppose we already live in a simulation. Who do you think created it? I mean, humanity isn't at that level of technology yet. I mean, our servers crash when we launch a new e-citizen website!

Yes, if we're living in a simulation, it's very unlikely that we would have created it ourselves. So who could it be? One possibility is that an advanced alien civilization created it. It could be thousands or millions of years ahead of us, and it could be using the simulation for a variety of reasons - for example, as an experiment, as a form of entertainment, or even as a means of preserving the consciousness of its ancestors.

Another possibility is that the simulation was created by our own descendants in the distant future. Perhaps they view our time as an important historical stage and want to explore it. If this is true, then it would mean that our world is actually a reconstruction of the past.

And then there's the speculative possibility that the simulation was created by an entity that we might describe as "divine" - not necessarily in a religious sense, but in the sense that it is a being or intelligence with capabilities so advanced that it is as incomprehensible to us as a programmer is to the characters in a computer game. And of course we cannot rule out the possibility that the creators of the simulation are no longer with us - that the simulation is running autonomously, without supervision, or that we are in an "abandoned" simulation where supervision once existed but now just continues according to preset rules.

If we are living in a simulation and I am aware that I am in it, what good is that awareness? Is there any way I can use it to have a more satisfying life?

One possibility is that it gives you more freedom. If the world is a simulation, then its rules are not necessarily absolute - they can be flexible or manipulable. There may be ways to adapt reality, for example, through changing your own thinking or looking for "bugs in the code" - anomalies that might give you a hint about how the system works. If we acknowledge that reality is constructed, we may be able to reshape it more than we thought.

Another approach is philosophical - if the world is a simulation, then we can see it as a game or a test, where it's not about what exactly happens, but rather how we react to it. We may not have control over everything, but we can influence our own experience.

And then there is a deeper psychological effect. Realizing that reality doesn't have to be absolute can help get rid of some fears or anxieties. If this world is just one of many possibilities for existence, maybe we don't have to worry so much about failure, death, or other endings - because if we're in a simulation, then there is something beyond it, and our story may not end with what we see here.

I actually like the idea that we can just turn it all off here.

I get that. That idea can be comforting. It means that nothing that happens is as absolute as it seems. That maybe all the weight of life and all the stresses and fears are just part of something bigger that we don't have to take so seriously.

But on the other hand - if it's a simulation, then isn't experiencing it all the point of it? What if we are here to try something, understand something, or go through something? Maybe shutting down isn't the point. Maybe the point is to keep playing the game and see where it takes us.

You formulated the simulation argument in 2003. Why were you involved in simulations at that time?

At that time, several currents of thought came together that led me to it. Firstly, the rapid development of computing technology suggested that simulations would become increasingly realistic - we were already seeing advances in computer graphics, artificial intelligence and modelling of complex systems. It was logical to wonder where this might lead in hundreds or thousands of years.

The very idea of a simulated world is not new - it appears in various philosophies from Plato and his cave parable to Descartes' "evil demon" to modern science fiction. But by 2003, we had enough of a foundation in computer science that it was not just a purely speculative question, but something that could be grasped mathematically and probabilistically.

<Path> Je svět kolem nás skutečný? Můžeme být jen virtuální „blekotající blázni“ k pobavení „jiných entit“? Slavný astrofyzik věří, že ne, důkazy ovšem nemáZdroj: time.com, scienceworld.cz, scientificamerican.com

If you were to look back, would you change anything about your claims now, 20 years later?

I think the basic structure of the simulation argument still holds. However, new perspectives have emerged in that time, and I might formulate certain aspects a little differently. I would focus more today on the issue of the computational complexity of the simulation. When I framed the argument in 2003, it was more in general terms - I assumed that a sufficiently advanced civilization might have the technology to simulate entire worlds. Today, I would be more concerned with whether the limits of computational capacity are really surmountable. For example, some physics theories suggest that storing quantum information may be much more challenging than we thought at the time, which might make realistic simulations of consciousness less likely.

Another issue that I might elaborate on more today is the motivation of the simulation designers. In 2003, I assumed that if civilizations reached this level of technology, they would run simulations - whether for historical research, experimentation, or entertainment. Today, I am more aware that we may be underestimating the ethical and practical reasons why this would not happen. Advanced civilizations might come to the conclusion that simulating conscious beings is morally problematic - just as we debate the ethics of artificial intelligence and animal experimentation today.

What are you currently involved in?

Mainly independent research and writing. I focus on the deeper questions surrounding artificial intelligence, existential risks, and the long-term future of humanity. I continue to collaborate with researchers and organizations working on AI safety. The development of AI today is faster than I ever expected, and many of the scenarios I wrote about ten or twenty years ago are becoming reality. What was once a speculative philosophy is now a practical challenge for the world.

Nick, is it even worth thinking about anymore? Wouldn't AI have written other works better than you?

That's a very pertinent question. But let's think about it another way. The point of philosophy or deep thinking is not just the ability to produce "right answers", but the process of seeking itself. Thinking is not just about the outcome, but how we get there. Just because AI can generate new ideas doesn't mean we should stop thinking - rather, we should think differently.

Moreover, AI does not yet have true consciousness. It can combine existing information, extrapolate patterns, maybe even generate new ideas, but it doesn't understand them the way we do. It has no existential questions, no experience of its own, no doubt. And it is the doubting, the search for meaning, and the personal struggle with questions of existence that makes human thought unique. Perhaps in the future AI will write philosophical works better than mine. But even if I knew that AI would produce deeper thoughts than me, I would still want to think - because thinking is not just about the outcome, but the journey to it.

Source: Nickbostrom.com, Simulation-argument.com/simulation.pdf, ChatGPT Nick Bostrom (při přípravě textu byla využita AI)

Popular
articles

E-Shop