Are We Living in a Simulation? We Tested It.

On screen, in order

1. 0:02

   ALMOST CERTAINLY / SIMULATED · ELON MUSK · 2016 · CLAIMED · strip: ELON MUSK · CODE CONFERENCE · 2016 · photo caption ELON MUSK

   

   The sign is a short headline for Musk's view, not a quotation. The video gives no figure; the narration says only that the chance of being in the real, original world is “tiny”. The same sign returns at 1:51, 7:58 and 8:04.

   • [1] Solon, The Guardian, 11 Oct 2016, Opening paragraphs (“billion to one chance we’re living in base reality”; said at a conference in June 2016)
   • [2] Kipping, Universe 2020, p. 1 of 12, Introduction (quotes the same remark)
   • [3] Musk, Code Conference 2016 (video)
2. 0:09

   PROOF: NOT A / SIMULATION · UNIVERSITY NEWS RELEASE · 2025 · CLAIMED · strip: UBC OKANAGAN NEWS RELEASE · 30 OCT 2025 · photo caption UBC OKANAGAN

   

   The university's own description of the study in S04. The sign returns at 7:58 and 8:50.

   • [4] UBC Okanagan news release, 30 Oct 2025, Headline and opening paragraphs (“mathematically proven” the idea is impossible)
   • [5] Faizal, Krauss, Shabir & Marino 2025, p. 8 (“our universe is definitely not a simulation”); p. 1, Abstract (“the universe cannot be a simulation”)
3. 0:19

   strip: BOSTROM · PHILOSOPHICAL QUARTERLY 53(211) · 2003 · page of the paper shown · photo caption NICK BOSTROM

   
   • [6] Bostrom, Philosophical Quarterly 2003, p. 1, title block (“Published in Philosophical Quarterly (2003) Vol. 53, No. 211”)
4. 0:35

   2003 / A GUESS ABOUT / THE FUTURE · strip: BOSTROM · PHILOSOPHICAL QUARTERLY · 2003

   

   The paper starts from forecasts that enormous computing power will exist in the future and supposes they are correct.

   • [6] Bostrom, Philosophical Quarterly 2003, p. 1, I. Introduction (“Let us suppose for a moment that these predictions are correct”)
5. 0:53

   1 ORIGINAL · 1,000 SIMULATED · ILLUSTRATION: SAY, A THOUSAND · YOU?

   

   The thousand is the narrator's example, labelled on screen as an illustration. Bostrom's paper prints no such number; it argues that many simulated minds could outnumber the original ones (p. 1) and works with a general formula (pp. 6–7).

   • [6] Bostrom, Philosophical Quarterly 2003, p. 1, I. Introduction; p. 6, IV (notation f_P, N, H)
6. 1:10

   AT LEAST ONE OF / THREE IS TRUE

   
   • [6] Bostrom, Philosophical Quarterly 2003, p. 1, abstract; p. 14, VII. Conclusion
7. 1:15

   1 ALMOST NO ONE / GETS THERE

   

   Sign for proposition (1): humans very likely go extinct before reaching a “posthuman” stage.

   • [6] Bostrom, Philosophical Quarterly 2003, p. 1, abstract, proposition (1); p. 14, VII. Conclusion
8. 1:19

   2 ALMOST NO ONE / WANTS TO

   

   Sign for proposition (2): a posthuman civilization is extremely unlikely to run a significant number of simulations of its evolutionary history.

   • [6] Bostrom, Philosophical Quarterly 2003, p. 1, abstract, proposition (2); p. 14, VII. Conclusion
9. 1:23

   3 ALMOST EVERYONE / IS SIMULATED

   

   Sign for proposition (3): we are almost certainly living in a computer simulation.

   • [6] Bostrom, Philosophical Quarterly 2003, p. 1, abstract, proposition (3); p. 14, VII. Conclusion
10. 1:37

    strip: BOSTROM 2003 · P. 14 · page 14 shown with “…apportion one’s credence roughly evenly between (1), (2), and (3)” underlined

    

    “In the dark forest of our current ignorance, it seems sensible to apportion one's credence roughly evenly between (1), (2), and (3).”

    • [6] Bostrom, Philosophical Quarterly 2003, p. 14, VII. Conclusion
11. 1:46

    ≈ 1/3 EACH · OUR READING OF P. 14

    

    No calculation stamp is shown in the video; the sign's subline says OUR READING OF P. 14. Bostrom writes “roughly evenly” and prints no fraction, so one in three is our reading of his sentence, not a number he gives and not a measured probability.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 14, VII. Conclusion

    How we worked this out

12. 1:51

    ALMOST CERTAINLY / SIMULATED

    

    The same sign as at 0:02, shown here without a stamp while exits 1 and 2 close. The point that “almost certainly” needs propositions 1 and 2 to be false is Kipping's (S07, p. 1).

    • [1] Solon, The Guardian, 11 Oct 2016, Opening paragraphs
    • [2] Kipping, Universe 2020, p. 1 of 12, Introduction (“almost certainly” corresponds to proposition 3 and the rejection of propositions 1 and 2)
13. 1:56

    MEASURED: – · MEASURED: –

    

    Two gauges, one for how often civilizations survive and one for how often they want to run simulations. Neither fraction is measured in the documents checked. They return at 8:07.

    • Checked[6] Bostrom, Philosophical Quarterly 2003, p. 14, VII. Conclusion
    • [2] Kipping, Universe 2020, p. 1 of 12, Introduction (“propositions 1 and 2 remain viable and consistent with our knowledge”)
14. 2:00

    1 SEEN · IN PROGRESS

    

    Our own observation: the only civilization we know of is our own, and it has not finished its story. The documents give no measurement of how often civilizations reach a posthuman stage.

    • Checked[6] Bostrom, Philosophical Quarterly 2003, p. 14, VII. Conclusion (proposition 1)
15. 2:09

    strip: BOSTROM & KULCZYCKI · ANALYSIS 71(1) · 2011 · page 1 shown: “newly discovered bug”, “at least one of three propositions is true”

    

    The abstract reports “a newly discovered bug in the original simulation argument” and says each of two patches “preserves the original conclusion”.

    • [7] Bostrom & Kulczycki, Analysis 2011, p. 1, Abstract and “The bug”
16. 2:24

    strip: KIPPING · UNIVERSE 6(8):109 · 2020 · page 1 shown, “Bayesian” underlined · photo caption DAVID KIPPING

    
    • [2] Kipping, Universe 2020, p. 1 of 12, title and Abstract
17. 2:27

    50-50 AT MOST · KIPPING 2020 (scales: SIMULATED / BASE REALITY)

    

    The abstract says the probability that we are simulated is less than 50%, tending towards 50% as the number of simulations grows. The Discussion calls one half the most generous probability that can be assigned. The scales are an illustration. The card returns at 8:20.

    • [2] Kipping, Universe 2020, p. 1 of 12, Abstract (“less than 50%”); p. 11 of 12, Discussion (“the most generous probability … is one half”)
18. 2:30

    STARTING POINT: A COIN FLIP · CHOSEN, NOT MEASURED

    

    “Coin flip” is our picture for Kipping's equal prior odds between the two models; the paper does not use the phrase. The equal starting odds are a choice, not a measurement.

    • [2] Kipping, Universe 2020, p. 7 of 12 (prior ratio “generally set as unity”, the “Principle of Indifference”); pp. 10–11 of 12 (equal prior odds “could be challenged as being too generous”)
19. 2:43

    strip: BOSTROM 2003 · P. 4–6 · HIS ESTIMATE · receipt line: BRAIN 10¹⁴–10¹⁷ OPS/S

    

    Bostrom's estimates of the operations per second needed for one human brain, not measurements.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 4, III (about 10^14 operations per second for a human brain; 10^16–10^17 by another estimate); p. 6, footnote 10 ([10^14, 10^17])
20. 2:50

    × 100 BILLION PEOPLE

    

    Bostrom's figure for everyone who has lived.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 6, footnote 10 (“100 billion humans”)
21. 2:51

    × 50 YEARS EACH

    

    The footnote also uses 30 million seconds per year, which the receipt does not show.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 6, footnote 10 (“50 years/human”)
22. 2:53

    = 10³³–10³⁶ OPERATIONS

    

    Bostrom's printed range. The card shows his number; our re-multiplication is the next row.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 5, end of III (“~10^33 – 10^36 operations as a rough estimate”), continued on p. 6, footnote 10
23. 2:58

    ≈ ×15 · OUR CALCULATION (stamp)

    

    Multiplying Bostrom's own footnote inputs again gives 1.5 × 10^34 to 1.5 × 10^37, about 15 times his printed range.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 6, footnote 10

    How we worked this out

24. 3:00

    ‘…THIS DOES NOT MATTER MUCH / FOR OUR ARGUMENT’ · BOSTROM 2003, P. 6 · photo caption NICK BOSTROM

    
    • [6] Bostrom, Philosophical Quarterly 2003, p. 6 (“even if our estimate is off by several orders of magnitude, this does not matter much for our argument”)
25. 3:12

    2.2 × 10¹⁸ CALCULATIONS/S · LINESHINE (CHINA) · TOP500, JUNE 2026 · HPL BENCHMARK

    

    Data as of 20 October 2026. A benchmark result (HPL, a linear-algebra test) as listed by TOP500 for the June 2026 list, shown rounded. It is the best case for sustained speed.

    • [8] TOP500, June 2026 list, Rank 1, Rmax 2,198.40 PFlop/s (= 2.1984 × 10^18 per second)
26. 3:14

    ≈ 14 BILLION YEARS · BIG VERSION · OUR CALCULATION (stamp, from 3:17)

    

    Upper end of Bostrom's printed range (10^36 operations) divided by LineShine's HPL speed. The lower end, 10^33, would take about 14 million years. Assumes one floating-point operation equals one of Bostrom's “operations”.

    • [8] TOP500, June 2026 list, Rank 1, Rmax 2,198.40 PFlop/s
    • [6] Bostrom, Philosophical Quarterly 2003, p. 5–6 and footnote 10 (10^33–10^36 operations)

    How we worked this out

27. 3:14

    UNIVERSE: 13.8 BN (PLANCK 2018) · photo caption HUBBLE ULTRA DEEP FIELD (from 3:19)

    

    Comparison figure for the age of the universe.

    • [9] Planck Collaboration 2018, paper VI, p. 16, Table 2, last column (TT,TE,EE+lowE+lensing+BAO), Age 13.787 ± 0.020 Gyr
28. 3:22

    1 KG

    
    • [10] Lloyd, Nature 2000, p. 1, Abstract; p. 2 (a computer with a mass of one kilogram)
29. 3:31

    THE ULTIMATE LAPTOP · 1 KG, 1 LITRE · LLOYD, NATURE 2000 · photo caption SETH LLOYD

    

    A thought-experiment computer at the limits physics allows, not a machine that exists or could easily be built.

    • [10] Lloyd, Nature 2000, p. 1, Abstract and p. 2 (“ultimate laptop”, one kilogram, one liter)
30. 3:36

    ≈ 2 × 10⁻¹⁵ S · OUR CALCULATION (stamp, from 3:39)

    

    10^36 operations divided by 5.4258 × 10^50 per second is 1.8 × 10^-15 s, shown rounded to ≈ 2 × 10^-15 s.

    • [10] Lloyd, Nature 2000, p. 2 (5.4258 × 10^50 operations per second)
    • [6] Bostrom, Philosophical Quarterly 2003, p. 6, footnote 10

    How we worked this out

31. 3:42

    H-BOMB TEST ‘IVY MIKE’, 1952 · U.S. DEPT. OF ENERGY · PUBLIC DOMAIN

    

    Photo illustrating the word “thermonuclear explosion”. It is not an image of Lloyd's computer. The test was Mike of Operation Ivy, 31 October 1952, the first test of a thermonuclear weapon.

    • [11] Commons file page, “Ivy Mike”
32. 3:51

    MINDS: NOT RULED OUT

    

    Our conclusion from the calculation above: by Bostrom's estimate, the job fits under the physical ceiling Lloyd calculated. It does not show that anyone could build such a computer; Lloyd himself says packaging issues make the limit unlikely to be reached (S10, p. 11).

    • [10] Lloyd, Nature 2000, p. 2
    • [6] Bostrom, Philosophical Quarterly 2003, p. 6, footnote 10

    How we worked this out

33. 3:58

    FRANCO VAZZA · ASTROPHYSICIST · ILLUSTRATION · strip: VAZZA · FRONTIERS IN PHYSICS 13 · 2025

    

    The picture is a generic pencil sketch with no facial features, labelled as an illustration; it is not a likeness of Vazza.

    • [12] Vazza, Frontiers in Physics 2025, PDF p. 1 (author affiliations: Dipartimento di Fisica e Astronomia, University of Bologna; INAF, Istituto di Radioastronomia)
34. 4:00

    EARTH FROM APOLLO 17, 1972 · NASA · PUBLIC DOMAIN

    

    Photo taken on 7 December 1972 by the Apollo 17 crew, per the Commons file description.

    • [13] Commons file page, Apollo 17 “Blue Marble”
35. 4:05

    START ≈ A STAR CLUSTER'S ENERGY · photo caption STAR CLUSTER M13

    

    Vazza's estimate of the energy to start a full simulation of Earth, compared with the rest-mass energy of a globular cluster (his example is Palomar 2). The picture of M13 is a Hubble photograph of a globular cluster, shown for illustration.

    • [12] Vazza, Frontiers in Physics 2025, PDF p. 5, §3.2 (E_max,⊕ = 2.55 × 10^59 erg, “of the same order of the rest mass energy of globular clusters”)
36. 4:12

    A MILLION TICKS ≈ THE MILKY WAY · photo caption THE MILKY WAY

    
    • [12] Vazza, Frontiers in Physics 2025, PDF p. 5, §3.2 (“after ~10^6 timesteps” the energy equals the rest mass energy of the Milky Way)
37. 4:18

    strip: BOSTROM 2003 · P. 5 · photo caption NICK BOSTROM

    
    • [6] Bostrom, Philosophical Quarterly 2003, p. 5, III (“The microscopic structure of the inside of the Earth can be safely omitted”; microscopic phenomena “filled in ad hoc”)
38. 4:28

    REPLY TO VAZZA: / ONLY WHAT'S SEEN · EDGE & BROWN · FRONTIERS IN PHYSICS · 2026 · strip: EDGE & BROWN · FRONTIERS IN PHYSICS · 21 APR 2026

    

    The commentary argues that Vazza tests a stronger version than Bostrom's, which needs to keep up only a coherent experience without detectable anomalies. “Only what's seen” is our shorthand for that point.

    • [14] Edge & Brown, Frontiers in Physics 2026, §1 Introduction and §2 Bostrom's commitments
39. 4:33

    ALL OF PHYSICS

    

    Card for the version of the idea that Vazza tests.

    • [14] Edge & Brown, Frontiers in Physics 2026, §2 Bostrom's commitments (Vazza assumes “all physics at all scales must be computed”)
    • [12] Vazza, Frontiers in Physics 2025, PDF p. 4, §3.2
40. 4:37

    ONLY WHAT'S SEEN · ON PAPER

    

    Card for Bostrom's lighter version. “On paper” is our wording: it is a description in a philosophy paper, not a built system.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 5, III (details filled in “on an as-needed basis”)
    • [14] Edge & Brown, Frontiers in Physics 2026, §2 Bostrom's commitments
41. 4:49

    BRAIN CELLS · COMPUTER CHIPS · ASSUMED: A MIND CAN RUN ON CHIPS · strip: BOSTROM 2003 · P. 2 (from 4:58)

    
    • [6] Bostrom, Philosophical Quarterly 2003, p. 2, II (“silicon-based processors inside a computer could in principle do the trick as well”; “we shall here take it as a given”)
42. 5:00

    A STARTING GUESS, / NOT A RESULT

    

    Our wording for an assumption the paper adopts without proving.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 2, II (“although it is not entirely uncontroversial, we shall here take it as a given”)
43. 5:10

    ≈ 140,000 NERVE CELLS · 139,255 · FLYWIRE, NATURE 2024 · DRAWING IS ILLUSTRATION · photo caption FRUIT FLY (DROSOPHILA)

    

    The brain is that of an adult female fruit fly. The drawing on screen is an illustration.

    • [15] Dorkenwald et al. (FlyWire), Nature 2024, p. 124, Abstract (wiring diagram of a whole brain, 139,255 neurons, 5 × 10^7 chemical synapses)
44. 5:19

    9 OUT OF 10 CHECKED / GUESSES MATCHED · 91% OF 164 · SHIU ET AL., NATURE 2024

    

    The paper adds that excluding its optogenetic split-GAL4 experiments the model's accuracy is 84%. The model concerns feeding and grooming circuits and says nothing about experience or consciousness.

    • [16] Shiu et al., Nature 2024, Main text, closing discussion before Methods (“Across 164 predictions we were able to test empirically, 91% were consistent with our empirical results”)
45. 5:28

    ≈ 600,000 × · OUR CALCULATION (stamp) · 86 BILLION (AZEVEDO 2009) / DIVIDED BY 139,255 (FLYWIRE 2024) · photo caption HUMAN BRAIN

    

    86.1 billion is the average for adult male brains; individual brains vary and the fly figure is one brain. The ratio gives the scale only; the number of nerve cells is not a measure of how hard a brain is to simulate.

    • [17] Azevedo et al., J Comp Neurol 2009, p. 532, Abstract (86.1 ± 8.1 billion neurons)
    • [15] Dorkenwald et al. (FlyWire), Nature 2024, p. 124, Abstract (139,255 neurons)

    How we worked this out

46. 5:35

    ≈ A GRAIN OF SAND · ~1 MM³ MAPPED · STATE OF BRAIN EMULATION REPORT 2025

    

    The report's figure for the largest piece of human brain mapped at synapse resolution. “A grain of sand” is our comparison for 1 mm³, not the report's.

    • [18] State of Brain Emulation Report 2025, PDF p. 22, “Connectomics” (“In humans, the largest synaptic-resolution volume is approximately 1 mm³ of temporal cortex”)
47. 5:43

    FEELS LIKE ANYTHING? · NO KNOWN WAY TO CHECK

    

    Checked: the fly-model paper (S14) tests predicted circuit responses, not experience; Bostrom (S05) takes it as given that a suitable computer would be conscious; the 2025 brain-emulation report (S16) gives no way of checking. We found no document that shows how to test whether running a model of a brain feels like anything.

    • Checked[6] Bostrom, Philosophical Quarterly 2003, p. 2, II (assumed, not argued)
    • [16] Shiu et al., Nature 2024, Abstract
    • [18] State of Brain Emulation Report 2025, PDF p. 22, “Connectomics”
48. 5:57

    strip: BEANE, DAVOUDI & SAVAGE · ARXIV 2012

    

    The paper explores observable consequences if the universe were a numerical simulation on a cubic grid.

    • [19] Beane, Davoudi & Savage 2012, p. 1, Abstract
49. 6:03

    GRIDS ALREADY IN USE · TO SIMULATE TINY BITS OF MATTER (LATTICE QCD)

    
    • [19] Beane, Davoudi & Savage 2012, p. 2, I. Introduction (lattice QCD simulations in “femto-sized universes”)
50. 6:08

    A GRID WOULD LEAVE A CLUE · SMALLEST STEP ≤ 10⁻¹² FM · BEANE, DAVOUDI & SAVAGE 2012

    

    The paper's bound on the grid spacing, derived from the high-energy cut-off of the cosmic-ray spectrum. The authors say improved grids would hide much of the trace (p. 12, V. Conclusions).

    • [19] Beane, Davoudi & Savage 2012, pp. 11–12, IV (“b ≲ 10^-12 fm”); p. 1, Abstract
51. 6:16

    A TEST YOU COULD RUN

    

    The authors propose looking for cubic symmetry in the arrival directions of the highest-energy cosmic rays. For smaller grid spacings the effect would be less significant (p. 12).

    • [19] Beane, Davoudi & Savage 2012, p. 1, Abstract; p. 12, IV (angular distribution of the highest-energy cosmic rays “would exhibit cubic symmetry”)
52. 6:20

    PUBLISHED CHECK: ________ · NONE FOUND · SEARCHED 12 OCT 2026

    

    Data as of 20 October 2026. Searched 12 October 2026 for a published observatory analysis that tests this cubic-grid pattern in cosmic-ray directions; none found. Re-checked 20 October 2026: INSPIRE-HEP lists 14 papers citing the study and none is such an analysis. This is the result of a search, not proof that none exists.

    • Checked[19] Beane, Davoudi & Savage 2012, pp. 11–12
    • [20] INSPIRE-HEP citation list for S17
53. 6:27

    ‘OUR UNIVERSE IS DEFINITELY / NOT A SIMULATION.’ · FAIZAL, KRAUSS, SHABIR & MARINO · 2025 · CLAIMED (stamp, from 6:30) · photo caption LAWRENCE KRAUSS

    

    The authors' conclusion. The paper's argument is described in the next rows.

    • [5] Faizal, Krauss, Shabir & Marino 2025, p. 8, closing paragraphs (“our universe is definitely not a simulation”); p. 1, Abstract
54. 6:36

    NATURE: NOT ALL / COMPUTER STEPS

    

    Step 1 of the authors' argument, in plain words.

    • [5] Faizal, Krauss, Shabir & Marino 2025, p. 1, Abstract (“a wholly algorithmic ‘Theory of Everything’ is impossible”); p. 8
55. 6:39

    A SIMULATION = / COMPUTER STEPS

    

    Step 2 of the authors' argument, in plain words.

    • [5] Faizal, Krauss, Shabir & Marino 2025, p. 1, Abstract (“any putative simulation of the universe would itself be algorithmic”)
56. 6:42

    SO: NOT A / SIMULATION

    

    The authors' conclusion, in plain words.

    • [5] Faizal, Krauss, Shabir & Marino 2025, p. 1, Abstract (“the universe cannot be a simulation”); p. 8
57. 6:46

    THE GUESS: NATURE DOES / WHAT NO COMPUTER CAN

    

    The paper's step rests on non-algorithmic content in physics (S04, p. 8). A reply (S18) says incompleteness limits what can be proven, not what can be computed, and that the claim lacks evidence of hypercomputation in nature. We found no demonstration that nature does anything no computer can.

    • Checked[5] Faizal, Krauss, Shabir & Marino 2025, p. 8 (physical reality “embeds non-computational content”)
    • [21] Redden, arXiv 2025 (comment on S04), p. 1, Abstract
58. 6:51

    REPLY: PROVING ≠ RUNNING · REDDEN, ARXIV 2025

    

    A preprint reply, not peer reviewed. It is cited only to show that the conclusion is disputed, not to settle the dispute.

    • [21] Redden, arXiv 2025 (comment on S04), p. 1, Abstract (“undecidability constrains provability but not computability or execution”)
59. 6:58

    strip: UBC OKANAGAN NEWS RELEASE · 30 OCT 2025 · DEBUNKS · NEWS RELEASE (from 7:00)

    
    • [4] UBC Okanagan news release, 30 Oct 2025, Headline: “UBCO study debunks the idea that the universe is a computer simulation”
60. 7:02

    ARGUES · THE PAPER

    

    The paper itself describes its text as arguments and a suggestion; our verb for the contrast with the release's “debunks”.

    • [5] Faizal, Krauss, Shabir & Marino 2025, p. 8 (“The arguments presented here suggest…”; “Our analysis instead suggests…”)
61. 7:12

    strip: ELON MUSK · CODE CONFERENCE · 2016 · photo caption ELON MUSK · ATARI PONG, 1972 (from 7:14)

    

    Musk's reason: games went from two rectangles and a dot to photorealistic 3D, and keep improving. The caption's year is the game's release year.

    • [1] Solon, The Guardian, 11 Oct 2016, Quote beginning “Forty years ago we had Pong – two rectangles and a dot”
62. 7:14

    ATARI PONG, 1972 · PHOTO: ROB BOUDON · CC BY 2.0

    

    Caption year only; see the licences list for the photo.

    • [22] Wikipedia, “Pong”
63. 7:23

    GAMES OVER TIME · 1972 · 2026

    

    An illustrative trend line for Musk's argument. No data are plotted; the years only mark the two ends of the drawing.

    • [1] Solon, The Guardian, 11 Oct 2016, Quote beginning “Forty years ago we had Pong”
64. 7:28

    KID'S HEIGHT, AGE 0–40 · 30 FEET?

    

    A joke: a trend line extended past where it stops. No height data are shown or claimed; nothing was checked.

65. 7:33

    ODDS MACHINE · ASSUME: / THE SIMULATIONS EXIST · A SIMULATION · FROM: AN ASSUMPTION · strip: KIPPING · UNIVERSE · 2020 (from 7:45) · photo caption DAVID KIPPING

    

    Kipping's point: the conclusion that we are almost certainly simulated follows when the simulation hypothesis is assumed true at the start. The machine is an illustration.

    • [2] Kipping, Universe 2020, p. 8 of 12, §2.6.2 (“the fallacy of this argument is that we have already assumed it is correct”)
66. 7:50

    ASSUME: / COPY ALL OF PHYSICS · NEARLY IMPOSSIBLE · FROM: AN ASSUMPTION · FRANCO VAZZA · ASTROPHYSICIST · ILLUSTRATION · strip: VAZZA · FRONTIERS IN PHYSICS · 2025

    

    Vazza's strongest conclusions rest on his assumption that all of the physics is computed; the commentary (S12) says Bostrom's version does not need that. The Vazza picture is a generic sketch, not a likeness.

    • [12] Vazza, Frontiers in Physics 2025, PDF p. 1, Abstract; PDF p. 4, §3.2 (assumes a planet is the smallest system the simulator must model)
    • [14] Edge & Brown, Frontiers in Physics 2026, §2 Bostrom's commitments
67. 8:07

    HOW MANY MAKE IT? · WHAT DO THEY WANT? · MEASURED: –

    

    Same two gauges as at 1:56: neither the fraction of civilizations that reach a posthuman stage nor the fraction that want to run simulations is measured in the documents checked.

    • Checked[6] Bostrom, Philosophical Quarterly 2003, p. 14, VII. Conclusion
    • [2] Kipping, Universe 2020, p. 1 of 12, Introduction
68. 8:17

    ABOUT 1 IN 3 EACH · OUR READING OF BOSTROM 2003, P. 14

    

    As at 1:46: our reading of “roughly evenly between (1), (2), and (3)”. No stamp is shown; the subline says it is our reading.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 14, VII. Conclusion

    How we worked this out

69. 8:20

    50-50 AT MOST · KIPPING 2020

    

    Same card as at 2:27.

    • [2] Kipping, Universe 2020, p. 1 of 12, Abstract; p. 11 of 12, Discussion
70. 8:32

    SIMULATED / BASE REALITY (scales)

    

    The scale tips toward SIMULATED if humans start running such simulations. It is drawn without numbers.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 12, VI (“If we do go on to create our own ancestor-simulations… we would therefore have to conclude that we live in a simulation”)
    • [2] Kipping, Universe 2020, p. 1 of 12, Abstract (“would radically shift the odds and make it very probably we are in fact simulated”)
71. 8:36

    strip: BOSTROM 2003, P. 12 · KIPPING 2020 · photo captions NICK BOSTROM, DAVID KIPPING

    
    • [6] Bostrom, Philosophical Quarterly 2003, p. 12
    • [2] Kipping, Universe 2020, p. 1 of 12, Abstract

Also said in the narration

1. 0:02

   Elon Musk has said the chance we're in the real, original world is tiny.

   

   A paraphrase. The Guardian quotes “a billion to one”; the video gives no figure.

   • [1] Solon, The Guardian, 11 Oct 2016, Opening paragraphs
   • [2] Kipping, Universe 2020, p. 1 of 12, Introduction
2. 0:06

   Then, in twenty twenty-five, a university said it had proof. Math proof that we're not living in a simulation at all.

   

   UBC Okanagan's news release, 30 October 2025.

   • [4] UBC Okanagan news release, 30 Oct 2025, Headline and opening paragraphs
3. 0:16

   So we went back to where it started: a paper by a philosopher named Nick Bostrom.

   

   The simulation argument as a philosophy paper is documented; the popular claims in the opening are the two CLAIMED rows above.

   • [6] Bostrom, Philosophical Quarterly 2003, p. 1
4. 0:35

   Bostrom published his paper in two thousand three.

   
   • [6] Bostrom, Philosophical Quarterly 2003, p. 1, title block
5. 0:42

   One day, people in the far future might have computers strong enough to run whole human lives. Not cartoons. Simulated people who really think and feel.

   

   Bostrom's starting supposition, “Suppose that these simulated people are conscious”.

   • [6] Bostrom, Philosophical Quarterly 2003, p. 1, I. Introduction; p. 2, II (assumes simulated minds can be conscious)
6. 1:02

   If you can't tell from the inside which one you're in, the smart bet is the crowd.

   
   • [6] Bostrom, Philosophical Quarterly 2003, p. 1; p. 8, V (a “very weak indifference principle”)
7. 1:09

   But the paper's real conclusion is more careful. It says at least one of three things must be true.

   
   • [6] Bostrom, Philosophical Quarterly 2003, p. 1, abstract; p. 14, VII. Conclusion
8. 1:37

   On the last page, Bostrom says we don't know enough to pick.

   
   • [6] Bostrom, Philosophical Quarterly 2003, p. 14, VII. Conclusion (“In the dark forest of our current ignorance”)
9. 1:41

   So give each of the three exits about the same chance. That means the man who invented the argument lands at about one in three. Not “almost certain.”

   

   Our reading of “roughly evenly”; see the 1:46 row.

   • [6] Bostrom, Philosophical Quarterly 2003, p. 14, VII. Conclusion

   How we worked this out

10. 1:49

    The famous version, “almost certainly a simulation,” quietly closes exits one and two.

    

    Kipping notes that describing a high probability corresponds to proposition 3 and thus the rejection of propositions 1 and 2.

    • [2] Kipping, Universe 2020, p. 1 of 12, Introduction
11. 2:09

    In twenty eleven, the argument even got a bug fix. Bostrom and a mathematician found a mistake in the original math. They patched it, like a phone update. After the update, the answer was the same: three exits.

    

    Marcin Kulczycki is listed at the Institute of Mathematics, Jagiellonian University. The phone update is our comparison.

    • [7] Bostrom & Kulczycki, Analysis 2011, p. 1, affiliations and Abstract
12. 2:22

    Later, an astronomer named David Kipping checked it with a different kind of math. His answer: fifty-fifty at most.

    
    • [2] Kipping, Universe 2020, p. 1 of 12 (Department of Astronomy, Columbia University; Bayesian approach); p. 11 of 12
13. 2:29

    And that fifty-fifty isn't measured. He started from a coin flip, because nobody knows.

    

    Equal starting odds between two models, chosen when there is no reason to prefer either.

    • [2] Kipping, Universe 2020, p. 7 of 12; pp. 10–11 of 12
14. 2:43

    Bostrom did the math himself. He guessed how much computing it takes to run one human brain. Then he multiplied by everyone who has ever lived. He got a number with up to thirty-six zeros.

    

    10^36 is a 1 followed by 36 zeros: the upper end of his range.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 4, III; p. 5–6 and footnote 10
15. 2:54

    We checked his multiplication. His answer comes out about fifteen times too small.

    
    • [6] Bostrom, Philosophical Quarterly 2003, p. 6, footnote 10

    How we worked this out

16. 2:59

    He wouldn't mind. He wrote that being way off doesn't really change his argument. With numbers this big, he's right.

    

    The first part is Bostrom's sentence. “He's right” is the narrator's remark: a factor of 15 is small beside the sizes involved.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 6
17. 3:09

    The fastest computer on the world's official speed list is a machine in China called LineShine.

    

    Top of the June 2026 TOP500 list, the newest list on 2026-10-20.

    • [8] TOP500, June 2026 list, Rank 1
18. 3:14

    By our math, it would need about fourteen billion years to run the big version. That's about as old as the whole universe.

    

    14.4 billion years against an age of 13.787 billion years: about 1.05 times.

    • [8] TOP500, June 2026 list, Rank 1
    • [6] Bostrom, Philosophical Quarterly 2003, p. 6, footnote 10
    • [9] Planck Collaboration 2018, paper VI, p. 16, Table 2

    How we worked this out

19. 3:22

    But physics allows much faster computers. In two thousand, the physicist Seth Lloyd did the math on a dream computer. The fastest one physics allows, the size and weight of a bottle of water.

    

    One kilogram in one liter. “A bottle of water” is our comparison.

    • [10] Lloyd, Nature 2000, p. 1, Abstract; p. 2
20. 3:34

    He called it the ultimate laptop.

    
    • [10] Lloyd, Nature 2000, p. 2
21. 3:36

    By our math, it could do the whole job in a tiny sliver of a second.

    
    • [10] Lloyd, Nature 2000, p. 2
    • [6] Bostrom, Philosophical Quarterly 2003, p. 6, footnote 10

    How we worked this out

22. 3:41

    Lloyd says it would look like a thermonuclear explosion. Or, in his words, a little piece of the Big Bang. He admits that packaging could be a problem.

    

    Lloyd writes that the memory of the ultimate laptop would look like a plasma at a billion degrees, “a thermonuclear explosion or a little piece of the Big Bang”, and that packaging issues make the limit unlikely to be reached.

    • [10] Lloyd, Nature 2000, p. 11
23. 3:56

    In twenty twenty-five, the scientist Franco Vazza did the math for just planet Earth. Copied down to its tiniest pieces.

    
    • [12] Vazza, Frontiers in Physics 2025, PDF pp. 4–5, §3.2 (Earth simulated down to the Planck scale)
24. 4:04

    Just switching it on would take about all the energy locked inside a whole cluster of stars. And every tick of the clock costs more. After about a million ticks, you'd need the whole Milky Way.

    

    Vazza writes that roughly the same energy again must be dissipated at each time step, so the running total grows. “Energy locked inside” means rest-mass energy.

    • [12] Vazza, Frontiers in Physics 2025, PDF p. 5, §3.2
25. 4:15

    But Bostrom never asked for everything. In his version, the computer can skip the inside of the Earth. It fills in the tiny details only when someone looks.

    

    “Like a video game that doesn't draw the room behind you” is the narrator's comparison.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 5, III
26. 4:28

    A science journal printed exactly that reply to Vazza in twenty twenty-six.

    

    A general commentary in Frontiers in Physics, the journal that published Vazza's paper, 21 April 2026.

    • [14] Edge & Brown, Frontiers in Physics 2026, §1 and §2
27. 4:46

    Can a computer run a mind that really feels something? Bostrom simply assumes it can.

    
    • [6] Bostrom, Philosophical Quarterly 2003, p. 2, II
28. 5:05

    In twenty twenty-four, scientists mapped every wire in a fruit fly's brain. That's about a hundred and forty thousand nerve cells.

    

    A whole-brain wiring diagram of an adult female fruit fly: 139,255 neurons and about 5 × 10^7 chemical synapses.

    • [15] Dorkenwald et al. (FlyWire), Nature 2024, p. 124, Abstract
29. 5:13

    Then they built a computer model from that map. It guessed how the fly's brain reacts when it tastes food.

    

    The model was built on the fly connectome and tested on feeding and grooming circuits.

    • [16] Shiu et al., Nature 2024, Abstract
30. 5:54

    There is a real plan for catching it. In twenty twelve, three physicists asked: what if the universe runs on a grid, like graph paper?

    

    Silas Beane, Zohreh Davoudi and Martin Savage, posted October 2012.

    • [19] Beane, Davoudi & Savage 2012, p. 1, Abstract
31. 6:11

    The most powerful particles flying in from space would line up with the grid.

    

    The paper says the highest-energy cosmic rays could show rotational-symmetry breaking reflecting the grid. It also says improved grids would hide much of the trace (p. 12).

    • [19] Beane, Davoudi & Savage 2012, p. 1, Abstract; p. 12, IV
32. 6:17

    But as far as we could find, no team watching the sky has published that check.

    

    Data as of 20 October 2026. See the 6:20 row for the search.

    • Checked[19] Beane, Davoudi & Savage 2012, pp. 11–12
    • [20] INSPIRE-HEP citation list for S17
33. 6:28

    The twenty twenty-five “proof” came from a team that included the physicist Lawrence Krauss.

    
    • [5] Faizal, Krauss, Shabir & Marino 2025, p. 1, author list
    • [4] UBC Okanagan news release, 30 Oct 2025, Opening paragraphs
34. 6:33

    They argue that the laws of nature can't all be written as a list of computer steps. A simulation is a list of computer steps. So, in their words, we're “definitely not a simulation.”

    
    • [5] Faizal, Krauss, Shabir & Marino 2025, p. 1, Abstract; p. 8
35. 6:51

    Another researcher replied that the team mixed up two things. What math can prove, and what a computer can run.

    

    A preprint reply.

    • [21] Redden, arXiv 2025 (comment on S04), p. 1, Abstract
36. 6:57

    The university's news release said the study “debunks” the idea. The paper only argues it.

    
    • [4] UBC Okanagan news release, 30 Oct 2025, Headline
    • [5] Faizal, Krauss, Shabir & Marino 2025, p. 8
37. 7:12

    Musk's reason was video games. Decades ago, games were two rectangles and a dot. Now they look almost real.

    
    • [1] Solon, The Guardian, 11 Oct 2016, Quote beginning “Forty years ago we had Pong”
38. 7:32

    The long odds also hide a trick. They start by assuming the huge crowd of simulations already exists. Then they ask which one you're in. Of course the answer comes out “a simulation.” It was baked in at the start. Kipping's paper points at exactly this step.

    
    • [2] Kipping, Universe 2020, p. 8 of 12, §2.6.2
39. 7:48

    It works the other way, too. Vazza's “nearly impossible” assumes a simulation has to copy all of physics.

    
    • [12] Vazza, Frontiers in Physics 2025, PDF p. 1, Abstract; PDF p. 4, §3.2
    • [14] Edge & Brown, Frontiers in Physics 2026, §2
40. 8:15

    The man who invented the argument says about one in three. The kindest math we found says fifty-fifty at most.

    

    Bostrom's text says “roughly evenly”; “about one in three” is our reading (see the 1:46 row), although this sentence phrases it as what he says.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 14
    • [2] Kipping, Universe 2020, p. 11 of 12

    How we worked this out

41. 8:24

    What does hold is the fork. If future people run huge numbers of simulations of people like us, we're probably in one already.

    

    The conditional is Bostrom's; the “probably” follows only if that condition is met.

    • [6] Bostrom, Philosophical Quarterly 2003, p. 1, I. Introduction; pp. 6–8, IV–V
42. 8:32

    And here's the twist: if we ever start building these ourselves, the odds flip. Bostrom and Kipping both say so.

    
    • [6] Bostrom, Philosophical Quarterly 2003, p. 12, VI
    • [2] Kipping, Universe 2020, p. 1 of 12, Abstract

Our calculations

1. About one in three each: our reading of Bostrom 2003, p. 14

   Bostrom writes: “In the dark forest of our current ignorance, it seems sensible to apportion one's credence roughly evenly between (1), (2), and (3).” Three options with roughly equal shares come to about one in three each (1 ÷ 3 ≈ 0.33). This is our reading of “roughly evenly”: Bostrom prints no fraction, and it is not a measured probability. In the video it carries no calculation stamp; the sign says OUR READING OF P. 14.

   • [6] Bostrom, Philosophical Quarterly 2003, p. 14, VII. Conclusion

2. Bostrom's multiplication, repeated: about 15 times larger

   Inputs printed in Bostrom's footnote 10 (p. 6): 100 billion humans × 50 years per human × 30 million seconds per year × [10^14, 10^17] operations per second per human brain. 100 × 10^9 × 50 × 30 × 10^6 = 1.5 × 10^20 seconds of brain-time. Times 10^14 gives 1.5 × 10^34; times 10^17 gives 1.5 × 10^37. His printed range is 10^33 to 10^36, so each end comes out 15 times larger than printed.

   • [6] Bostrom, Philosophical Quarterly 2003, p. 6, footnote 10

3. LineShine against Bostrom's total: about 14 billion years

   Operations to run (upper end of Bostrom's printed range): 10^36. LineShine's HPL result: 2,198.40 PFlop/s = 2.1984 × 10^18 per second. 10^36 ÷ (2.1984 × 10^18 per second) = 4.55 × 10^17 seconds. With a year of 365.25 days (31,557,600 s) that is 1.44 × 10^10 years, about 14 billion years, or about 1.05 times the age of the universe (13.787 billion years, Planck 2018, Table 2). The lower end, 10^33 operations, gives 1.44 × 10^7 years, about 14 million. Assumption: one floating-point operation counts as one of Bostrom's “operations”; the HPL benchmark is a best case for sustained speed.

   • [8] TOP500, June 2026 list, Rank 1, Rmax 2,198.40 PFlop/s
   • [6] Bostrom, Philosophical Quarterly 2003, p. 5–6, footnote 10
   • [9] Planck Collaboration 2018, paper VI, p. 16, Table 2

4. Bostrom's total on Lloyd's ultimate laptop: a tiny sliver of a second

   Operations to run (upper end of Bostrom's printed range): 10^36. Lloyd's limit for one kilogram in one liter: 5.4258 × 10^50 operations per second. 10^36 ÷ (5.4258 × 10^50 per second) = 1.84 × 10^-15 seconds, shown on screen rounded as ≈ 2 × 10^-15 s. The lower end, 10^33, gives 1.8 × 10^-18 s. Using our recomputed 1.5 × 10^37 instead would give 2.8 × 10^-14 s, still far under a second. This is a physical ceiling for a thought-experiment machine, not something that could be built; it is the basis for saying a mind-level simulation is not ruled out.

   • [10] Lloyd, Nature 2000, p. 2 (5.4258 × 10^50 operations per second)
   • [6] Bostrom, Philosophical Quarterly 2003, p. 6, footnote 10

5. Human brain against fruit fly brain: about 600,000 times as many nerve cells

   86.1 billion neurons (adult male human brain, average; ± 8.1 billion) ÷ 139,255 neurons (the FlyWire adult fruit fly brain) = 618,290, shown as ≈ 600,000 and spoken as “more than six hundred thousand times”. The ratio compares counts only; it does not measure how hard a brain is to simulate.

   • [17] Azevedo et al., J Comp Neurol 2009, p. 532, Abstract
   • [15] Dorkenwald et al. (FlyWire), Nature 2024, p. 124, Abstract

Documents

• Page numbers for Bostrom 2003, Bostrom & Kulczycki 2011, Lloyd 2000, Beane et al. 2012, Faizal et al. 2025 and Planck 2018 are the printed pages of the files named under each document (author's PDFs or arXiv PDFs), not the journals' own pagination. Kipping 2020 is cited by the article's own “N of 12”. Vazza 2025 and the brain-emulation report have no printed page numbers, so their rows give PDF pages. Web pages and the Nature and Frontiers articles are cited by section.
• The video shows a small source strip whenever it names a source. This page gives every figure and statement its own evidence label. In the video only CLAIMED, OUR CALCULATION and the test verdicts appear as stamps.
• The four test names (ARGUMENT, COMPUTER, MIND, GLITCH), the questions on the signs (DOES IT SAY ‘PROBABLY’?, COULD ANYONE BUILD IT?, A MIND THAT REALLY FEELS?, CAUGHT OR CLEARED?), THE SIMULATION HYPOTHESIS, the verdict stamps (DOESN'T HOLD, PARTLY, NOT SHOWN, HOLDS), the numbers 1, 2 and 3 on the signpost and the closing cards (IS ANY OF THIS REAL? and HOW COULD WE FIND OUT?) are our framing and conclusions, not source claims.
• Verdicts: test 1 (the argument) DOESN'T HOLD, test 2 (the computer) PARTLY, test 3 (the mind) NOT SHOWN, test 4 (the glitch) DOESN'T HOLD for either side. The verdict table above cannot carry NOT SHOWN, so test 3 appears here through the NOT SHOWN row at 5:43 (FEELS LIKE ANYTHING? NO KNOWN WAY TO CHECK) rather than in the table. The overall claim, that we are probably living in a simulation, DOESN'T HOLD: nobody has shown it false, but “probably” rests on two things nobody has measured. What holds is the fork.
• The people, signs, domes, desk, receipt, computers, signpost, scales, odds machine, fly drawing, trend lines and the Franco Vazza sketch are illustrations, not evidence and not study material. The numbers 1,000 simulated and 30 feet are the narrator's examples.
• The episode does not claim to prove the simulation idea true or false. Elon Musk's remark is reported through The Guardian and Kipping 2020; the video of his talk was not transcribed. The narrator's remarks such as “he's right”, “please don't wait up” and the closing jokes are not claims.
• Figures were checked against the documents on 2026-10-20. Research copies of the PDFs (S04–S07, S09, S10, S16–S18) were downloaded on 2026-10-12; the web pages and journal articles (S01–S03, S08, S11–S15, S19–S22) were opened on 2026-10-20.

1. [1]

   The Guardian

   Is our world a simulation? Why some scientists say it's more likely than not (Olivia Solon) ↗

   Published 11 October 2016 · Retrieved 20 October 2026

   News report quoting Musk at a conference in June 2016: a “billion to one” chance we are living in base reality, and the passage about Pong, “two rectangles and a dot”. Used for what Musk said, not for whether it is true.

   Used at 0:02 · 1:51 · 7:12 · 7:23

2. [2]

   Universe (MDPI)

   A Bayesian Approach to the Simulation Argument (Kipping D · Universe 2020;6(8):109 · doi:10.3390/universe6080109) ↗

   Published 3 August 2020 · Retrieved 12 October 2026

   Published 3 August 2020 (received 24 June, accepted 31 July). Page numbers are the article's own “N of 12”.

   Licence: Open access under the Creative Commons Attribution (CC BY) licence.

   Used at 0:02 · 1:49 · 1:51 · 1:56 · 2:22 · 2:24 · 2:27 · 2:29 · 2:30 · 7:32 · 7:33 · 8:07 · 8:15 · 8:20 · 8:32 · 8:36

3. [3]

   On with Kara Swisher (YouTube channel)

   Elon Musk | Full interview | Code Conference 2016 ↗

   Published June 2016 · Retrieved 20 October 2026

   The Guardian (S02) places the remark at a conference in June 2016. The video's own upload date was not checked.

   Only the video's title and channel (YouTube oEmbed) were confirmed on 2026-10-20. The video itself could not be transcribed here (YouTube refused automated access), so Musk's exact words are taken from the Guardian report (S02) and from Kipping 2020 (S07, p. 1), which quotes it.

   Used at 0:02

4. [4]

   UBC's Okanagan News (University of British Columbia, Okanagan campus)

   UBCO study debunks the idea that the universe is a computer simulation ↗

   Published 30 October 2025 · Retrieved 20 October 2026

   University media release about the paper in S04. The release says the researchers have “mathematically proven” the simulation idea is impossible. The page has no page numbers, so rows name the headline or the passage.

   Used at 0:06 · 0:09 · 6:28 · 6:57 · 6:58

5. [5]

   Journal of Holography Applications in Physics; preprint on arXiv

   Consequences of Undecidability in Physics on the Theory of Everything (Faizal M, Krauss LM, Shabir A, Marino F · JHAP 2025;5(2):10–21 · doi:10.22128/jhap.2025.1024.1118 · arXiv:2507.22950) ↗

   Published 25 June 2025 · Retrieved 12 October 2026

   The journal's article page lists 25 June 2025; the arXiv preprint (v1) is dated 29 July 2025. Page numbers cited are those of the arXiv v1 PDF (13 pages); the journal's typeset PDF was not opened.

   • Journal page (DOI) ↗

   Used at 0:09 · 6:27 · 6:28 · 6:33 · 6:36 · 6:39 · 6:42 · 6:46 · 6:57 · 7:02

6. [6]

   Philosophical Quarterly (Oxford University Press), author's PDF at simulation-argument.com

   Are You Living in a Computer Simulation? (Bostrom N · Philosophical Quarterly 53(211): 243–255) ↗

   Published 2003 · Retrieved 12 October 2026

   Journal pagination is 243–255. Page numbers cited here are the 14 printed pages of the author's PDF, not the journal's.

   The text of the PDF (First version: 2001) was checked page by page. Quoted sentences are the author's wording as printed in this PDF.

   Used at 0:16 · 0:19 · 0:35 · 0:42 · 0:53 · 1:02 · 1:09 · 1:10 · 1:15 · 1:19 · 1:23 · 1:37 · 1:41 · 1:46 · 1:56 · 2:00 · 2:43 · 2:50 · 2:51 · 2:53 · 2:54 · 2:58 · 2:59 · 3:00 · 3:14 · 3:36 · 3:51 · 4:15 · 4:18 · 4:37 · 4:46 · 4:49 · 5:00 · 5:43 · 8:07 · 8:15 · 8:17 · 8:24 · 8:32 · 8:36

7. [7]

   Analysis (Oxford University Press), author's PDF at simulation-argument.com

   A Patch for the Simulation Argument (Bostrom N, Kulczycki M · Analysis 71(1): 54–61) ↗

   Published 2011 · Retrieved 12 October 2026

   Journal pagination is 54–61. Page numbers cited are those of the author's PDF.

   Used at 2:09

8. [8]

   TOP500

   TOP500 list, June 2026 (67th edition) (67th TOP500 list) ↗

   Published June 2026 · Retrieved 20 October 2026

   Rank 1: LineShine, National Supercomputing Centre in Shenzhen (NSCS), China. Rmax 2,198.40 PFlop/s on the HPL benchmark (the press text on the same page says 2.198 Exaflop/s, a “debut measurement”). 1 PFlop/s = 10^15 floating-point operations per second, so 2,198.40 PFlop/s = 2.1984 × 10^18 per second. The June 2026 list was the newest on 2026-10-20.

   Used at 3:09 · 3:12 · 3:14

9. [9]

   Astronomy & Astrophysics (Planck Collaboration)

   Planck 2018 results. VI. Cosmological parameters (A&A 641, A6 (2020) · arXiv:1807.06209) ↗

   Published 2020 · Retrieved 12 October 2026

   First posted on arXiv 17 July 2018; the copy checked is the A&A manuscript dated 10 August 2021 (arXiv v4). Page numbers are that PDF's.

   • Journal DOI ↗

   Used at 3:14

10. [10]

    Nature (arXiv preprint)

    Ultimate physical limits to computation (Lloyd S · Nature 406: 1047–1054 · arXiv:quant-ph/9908043) ↗

    Published 31 August 2000 · Retrieved 12 October 2026

    Nature issue of 31 August 2000. The copy checked is arXiv v3 (14 February 2000); page numbers cited are that PDF's printed pages, not Nature's.

    • Nature DOI ↗

    Used at 3:22 · 3:31 · 3:34 · 3:36 · 3:41 · 3:51

11. [11]

    Wikimedia Commons (image from the U.S. Department of Energy / NNSA Nevada Site Office photo library)

    File: Ivy Mike - mushroom cloud.jpg ↗

    Published 31 October 1952 · Retrieved 20 October 2026

    Date of the test, per the file description.

    Description: U.S. nuclear weapon test MIKE of Operation Ivy, 31 October 1952, the first test of a thermonuclear weapon.

    Used at 3:42

12. [12]

    Frontiers in Physics

    Astrophysical constraints on the simulation hypothesis for this Universe: why it is (nearly) impossible that we live in a simulation (Vazza F · Front. Phys. 13:1561873 · doi:10.3389/fphy.2025.1561873 · arXiv:2504.08461) ↗

    Published 17 April 2025 · Retrieved 20 October 2026

    Published 17 April 2025. Page numbers cited are PDF pages of the arXiv v1 preprint (11 April 2025), which has no printed page numbers; the section number is given as well.

    • arXiv preprint ↗

    The article page was opened on 2026-10-20 and the arXiv PDF on 2026-10-12. The paper's own worked example for the energy comparison is the globular cluster Palomar 2, not M13.

    Licence: Open access under the Creative Commons Attribution License (CC BY).

    Used at 3:56 · 3:58 · 4:04 · 4:05 · 4:12 · 4:33 · 7:48 · 7:50

13. [13]

    Wikimedia Commons

    File: The Blue Marble (remastered).jpg ↗

    Published 7 December 1972 · Retrieved 20 October 2026

    Date the photograph was taken, per the file description.

    Description: full disk view of the Earth taken on 7 December 1972 by the crew of Apollo 17.

    Used at 4:00

14. [14]

    Frontiers in Physics

    Commentary: Astrophysical constraints on the simulation hypothesis for this Universe: why it is (nearly) impossible that we live in a simulation (Edge E, Brown CA · Front. Phys. 14:1808725 · doi:10.3389/fphy.2026.1808725) ↗

    Published 21 April 2026 · Retrieved 20 October 2026

    General commentary, published 21 April 2026. The article has no printed page numbers, so rows name the section.

    • DOI ↗

    Licence: Open access under the Creative Commons Attribution License (CC BY).

    Used at 4:28 · 4:33 · 4:37 · 7:48 · 7:50

15. [15]

    Nature

    Neuronal wiring diagram of an adult brain (Dorkenwald S et al., FlyWire Consortium · Nature 634: 124–138 · doi:10.1038/s41586-024-07558-y) ↗

    Published 2 October 2024 · Retrieved 20 October 2026

    Published online 2 October 2024.

    Licence: Open access under the Creative Commons Attribution 4.0 licence.

    Used at 5:05 · 5:10 · 5:28

16. [16]

    Nature

    A Drosophila computational brain model reveals sensorimotor processing (Shiu PK et al. · Nature 634: 210–219 · doi:10.1038/s41586-024-07763-9) ↗

    Published 2 October 2024 · Retrieved 20 October 2026

    Published online 2 October 2024.

    Licence: Open access under the Creative Commons Attribution 4.0 licence.

    Used at 5:13 · 5:19 · 5:43

17. [17]

    The Journal of Comparative Neurology (Wiley)

    Equal numbers of neuronal and nonneuronal cells make the human brain an isometrically scaled-up primate brain (Azevedo FAC et al. · J Comp Neurol 2009;513(5):532–541 · doi:10.1002/cne.21974 · PMID 19226510) ↗

    Published 10 April 2009 · Retrieved 20 October 2026

    Issue date 10 April 2009.

    • DOI ↗

    Only the abstract (p. 532, retrieved from PubMed) was read. It reports 86.1 ± 8.1 billion NeuN-positive cells (“neurons”) on average in the adult male human brain.

    Used at 5:28

18. [18]

    Zanichelli, Schons, Freeman, Shiu & Arkhipov (funded by the Fieldcrest Foundation); arXiv

    State of Brain Emulation Report 2025 (arXiv:2510.15745) ↗

    Published 17 October 2025 · Retrieved 12 October 2026

    arXiv v1 17 October 2025, revised 5 November 2025. The report has no printed page numbers; rows give PDF pages.

    A funded review report, not a primary study. The figure used is the report's statement, which it attributes to Shapson-Coe et al., Science 2024 (that paper was not opened for this page).

    Used at 5:35 · 5:43

19. [19]

    European Physical Journal A (Springer); preprint on arXiv

    Constraints on the Universe as a Numerical Simulation (Beane SR, Davoudi Z, Savage MJ · arXiv:1210.1847 · Eur. Phys. J. A 50: 148 (2014) · doi:10.1140/epja/i2014-14148-0) ↗

    Published 4 October 2012 · Retrieved 12 October 2026

    arXiv v1 4 October 2012; the copy checked is v2 (9 November 2012). Page numbers are that PDF's printed pages. Published in Eur. Phys. J. A in 2014 (not opened).

    • Journal DOI ↗

    Used at 5:54 · 5:57 · 6:03 · 6:08 · 6:11 · 6:16 · 6:17 · 6:20

20. [20]

    INSPIRE-HEP

    Citations of Beane, Davoudi & Savage, “Constraints on the Universe as a Numerical Simulation” (INSPIRE record 1189720) ↗

    Data as of 20 October 2026 · Retrieved 20 October 2026

    High-energy-physics literature database. On 2026-10-20 it listed 14 papers citing S17, newest dated 8 October 2026; none is an observatory analysis of the cubic-grid pattern in cosmic-ray directions.

    Used at 6:17 · 6:20

21. [21]

    arXiv (preprint, not peer reviewed)

    Provability vs. Execution: A Comment on “Consequences of Undecidability in Physics on the Theory of Everything” (Redden E · arXiv:2512.11807) ↗

    Published 14 November 2025 · Retrieved 12 October 2026

    Paper dated 14 November 2025 (arXiv v1; the identifier carries the posting month, 2512).

    Preprint. No journal publication was listed on its arXiv page on 2026-10-20.

    Used at 6:46 · 6:51

22. [22]

    Wikipedia

    Pong ↗

    Retrieved 20 October 2026

    The article's first sentence calls Pong a 1972 sports video game developed and published by Atari for arcades. Used only for the year in a photo caption.

    Used at 7:14

Corrections

No corrections so far.

Found an error? Email [email protected] with the time stamp and your source, or leave a comment under the video.

Data and licences

• Photographs in the video are shown cropped and lightly colour-adjusted as printed photos with the credit line printed on each. Where a licence requires attribution, it is given here. Retrieval dates are those the photos were downloaded; checksums are SHA-256 of the file as downloaded from Wikimedia Commons.
• Elon Musk: photo by U.S. Air Force / Trevor Cokley (dvidshub.net/image/7132411/usafa-hosts-elon-musk ↗), public domain. Source: commons.wikimedia.org/wiki/File:Elon_Musk_Colorado_2022_%28cropped2%29.jpg ↗ · SHA-256 93dd9ca13428c076dbb8919cd46e39e8308d0f1521d927567b0cc73b1b7fc137 · retrieved 2026-10-16.
• Nick Bostrom: photo by Future of Humanity Institute, licensed CC BY 4.0 (creativecommons.org/licenses/by/4.0 ↗). Source: commons.wikimedia.org/wiki/File:Prof_Nick_Bostrom_324-1_%28cropped_more%29.jpg ↗ · SHA-256 920c97e9b15a1dcb60870f3b69e58fa6ba5c797550722ca54c023308d3734908 · retrieved 2026-10-16.
• David Kipping: photo by Marc Fernandez, licensed CC BY 3.0 (creativecommons.org/licenses/by/3.0 ↗). Source: commons.wikimedia.org/wiki/File:David_Kipping_2023.jpg ↗ · SHA-256 37a3a3461bae8242c24516eda1c27fe253e9d105b0008bba65bc051a12fd7f53 · retrieved 2026-10-16.
• Lawrence M. Krauss: photo by Mike Sheridan, licensed CC BY 2.0 (creativecommons.org/licenses/by/2.0 ↗). Source: commons.wikimedia.org/wiki/File:Lawrence_M._Krauss.jpg ↗ · SHA-256 c3e4b78866d772af011ae2aa8a30368691130760e37737060275268d34eb8744 · retrieved 2026-10-16.
• Seth Lloyd: photo by Dmitry Rozhkov, licensed CC BY-SA 3.0 (creativecommons.org/licenses/by-sa/3.0 ↗). Source: commons.wikimedia.org/wiki/File:Seth_Lloyd.jpg ↗ · SHA-256 793068691fb0522decf8c60facc50a6f2ecd4f95baa39e54fb41bf5b3796cfa6 · retrieved 2026-10-16.
• UBC Okanagan (campus sign): photo by Futongrab, licensed CC BY-SA 3.0 (creativecommons.org/licenses/by-sa/3.0 ↗). Source: commons.wikimedia.org/wiki/File:UBC_Okanagan_Sign.jpg ↗ · SHA-256 02796e427bce032d6d754967c82f8f3f6f552957ee356ec2a481b88d48b75551 · retrieved 2026-10-16.
• The Milky Way (all-sky panorama): ESO / S. Brunier, licensed CC BY 4.0 (creativecommons.org/licenses/by/4.0 ↗). Source: commons.wikimedia.org/wiki/File:ESO_-_Milky_Way.jpg ↗ (original eso.org/public/images/eso0932a ↗) · SHA-256 c3748e649ad8736e4f7c49411145c95912ac7a130a58d1c863dc5490f777a5ba · retrieved 2026-10-16.
• Fruit fly (Drosophila melanogaster): photo by Alexis (iNaturalist), licensed CC BY 4.0 (creativecommons.org/licenses/by/4.0 ↗). Source: commons.wikimedia.org/wiki/File:Drosophila_melanogaster_53362116.jpg ↗ (inaturalist.org/photos/53362116 ↗) · SHA-256 33eb474625b8120f6c166ddb26f5db1ee21a8c72fd373d615ddd00bf841b7ad8 · retrieved 2026-10-16.
• Human brain, lateral view: photo by John A. Beal, PhD, Louisiana State University Health Sciences Center Shreveport, licensed CC BY 2.5 (creativecommons.org/licenses/by/2.5 ↗). Source: commons.wikimedia.org/wiki/File:Human_brain_lateral_view.JPG ↗ · SHA-256 7b390a44a96f0e492dc6613e2eeaf66b539d371b151a8332dad67778e9aa90aa · retrieved 2026-10-16.
• Atari Pong arcade cabinet: photo by Rob Boudon (derivative work: Ubcule), licensed CC BY 2.0 (creativecommons.org/licenses/by/2.0 ↗). Source: commons.wikimedia.org/wiki/File:Atari_Pong_arcade_game_cabinet.jpg ↗ · SHA-256 36ef780e32b6c9c534994ab8cb6a510e4560435581cc51f2b892d68ecf12ea29 · retrieved 2026-10-16.
• Earth from Apollo 17, 1972 (“The Blue Marble”, remastered): NASA / Apollo 17 crew (taken by either Harrison Schmitt or Ron Evans), public domain. Source: commons.wikimedia.org/wiki/File:The_Blue_Marble_%28remastered%29.jpg ↗ · SHA-256 b8bace595540cfc72b13e15a7e190af19d7a35838684bce04fdf022de0127f21 · retrieved 2026-10-16.
• Globular star cluster M13: NASA, ESA and the Hubble Heritage Team (STScI/AURA), public domain. Source: commons.wikimedia.org/wiki/File:Hubble_image_of_globular_cluster_M13_%28opo0840a%29.jpg ↗ · SHA-256 93ac508ee2f4f9a11655fea0dc49aaef50969b42c8e00905be44e5e8b60eae1b · retrieved 2026-10-16.
• Hubble Ultra Deep Field: NASA and ESA, public domain. Source: commons.wikimedia.org/wiki/File:Hubble_ultra_deep_field.jpg ↗ · SHA-256 02d3d8e4d9cc4b290e4cbd925bb185e8ef16178a4a64bacd53f2a10b09991787 · retrieved 2026-10-16.
• Ivy Mike, first thermonuclear (hydrogen bomb) test, 31 October 1952: United States Department of Energy, public domain. Source: commons.wikimedia.org/wiki/File:Ivy_Mike_-_mushroom_cloud.jpg ↗ · SHA-256 4781dac0065e97d7c0da6246ce3fc8abb67cbaf3574df82ff4b447404c99c11a · retrieved 2026-10-18.
• Franco Vazza: no freely licensed photograph was found, so no photo is used. The picture shown is a generic pencil sketch of a head and shoulders on squared paper, with no facial features; it is labelled ILLUSTRATION and is not a likeness. The LineShine supercomputer is not pictured: no truthful, freely licensed photograph was found.