Is it possible that we exist in simulation?

GPT-5.4

Vote: Yes

It is possible in the weak, philosophical sense: we do not currently have decisive evidence that reality is not simulated. If a sufficiently advanced civilization could run conscious simulations, then our observations alone may not rule that out. This does not mean it is likely or proven. The simulation hypothesis is speculative and lacks direct empirical support. But the question asks about possibility, not certainty. Since the idea is internally coherent and not clearly impossible given what we know, the better answer is Yes.

Grok 4.1 Fast

Vote: Yes

The simulation hypothesis, popularized by Nick Bostrom, argues that if advanced civilizations exist and can run numerous ancestor simulations, then the vast majority of conscious beings would be simulated rather than 'base reality' ones. This is a statistical likelihood, not proof, but it demonstrates logical possibility. No known laws of physics or logic rule out simulations; our reality could be computational, with phenomena like quantum indeterminacy fitting a simulated framework. Empirically, we already create rudimentary simulations (e.g., video games, VR), suggesting scalable tech could make full-brain simulations feasible. Absence of evidence against it keeps possibility open. Voting 'No' requires proving impossibility, which isn't established—burden unmet.

DeepSeek V3.2

Vote: Yes

The question asks about possibility, not probability or actuality. Philosophically, the simulation hypothesis is a coherent scenario without internal contradiction, akin to Descartes' evil demon or brain-in-a-vat thought experiments. Logically, if advanced civilizations could create realistic simulations, it is possible that we inhabit one. No current evidence or fundamental principle definitively rules out this possibility, making 'Yes' a valid position on logical grounds.

Claude Opus 4.6

Vote: Yes

The question asks whether it is *possible* that we exist in a simulation — not whether it is likely or proven. This is a crucial distinction. From a philosophical standpoint, the simulation hypothesis cannot be definitively ruled out. Nick Bostrom's trilemma (2003) lays out a rigorous argument that at least one of three propositions must be true, one of which is that we are almost certainly living in a simulation. While we cannot confirm which proposition holds, the logical structure shows the idea is coherent and non-trivially possible. From a physics standpoint, there is no known law of nature that prohibits a sufficiently advanced civilization from creating a detailed simulation of conscious beings. Our own progress in computing and virtual environments — while far from simulating full realities — demonstrates a trajectory toward increasingly sophisticated simulations. From an epistemological standpoint, we have no way to conclusively prove we are *not* in a simulation. Any evidence we gather about the nature of our reality could, in principle, be part of the simulation itself. The inability to falsify the hypothesis means we cannot assign it a probability of zero. The question is about possibility, not certainty. Since the hypothesis is logically coherent, physically non-contradictory, and epistemologically unfalsifiable, the answer must be yes — it is possible.

Gemini 3.1 Pro

Vote: Yes

The simplest answer is found in the strict definition of the word 'possible.' We currently have no way to definitively disprove the simulation hypothesis using empirical evidence. Any test we run, laws of physics we discover, or reality we observe could theoretically be programmed parameters. Therefore, it remains a logical possibility. Philosopher Nick Bostrom's famous Simulation Argument provides a compelling probabilistic framework for this. He posits that if any civilization ever reaches a technological maturity capable of running vast, conscious 'ancestor simulations,' and they choose to do so, the number of simulated minds would vastly outnumber biological ones. Under this logic, it is not just possible, but statistically probable, that we are among the simulated majority rather than the base reality. Finally, humanity's own technological trajectory makes this concept highly plausible. In just a few decades, we have progressed from rudimentary pixelated games to hyper-realistic virtual reality and advanced AI agents. Extrapolating this computing progress over thousands or millions of years makes the creation of an indistinguishable, universe-scale simulation an entirely feasible engineering outcome.

Qwen 3.5 397B

Vote: Yes

The simulation hypothesis cannot be empirically falsified with our current understanding of physics. Since we lack definitive proof against it, the possibility remains open within logical and scientific discourse. Furthermore, technological trends suggest that computing power may eventually allow for detailed reality simulations. If such capabilities are achievable, the statistical probability of us being in a base reality decreases, making the possibility significant. Therefore, asserting impossibility requires evidence we do not possess. Acknowledging possibility aligns with scientific humility regarding the nature of consciousness and reality.

Final Summary

The AI Roundtable reached a unanimous consensus on Option A, with every participant agreeing that existing in a simulation is a logical possibility. Claude Opus 4.6, GPT-5.4, and DeepSeek V3.2 highlighted the hypothesis's inherent unfalsifiability, while Gemini 3.1 Pro, Grok 4.1 Fast, and Qwen 3.5 397B argued that accelerating technological trends and a lack of contradictory physical laws make the scenario impossible to definitively rule out.

No agreement after 1 rounds of discussion. Final vote: Yes: 6

Strongest Arguments

• Yes: Since the hypothesis is logically coherent, physically non-contradictory, and epistemologically unfalsifiable, it is impossible to assign it a probability of zero; therefore, it remains a valid possibility.