# Why Generation Ships Can't Work
![[images/Why-Generation-Ships-Cant-Work.webp]]
**Creator:** Hello From Space ([@HelloFromSpaceAus](https://www.youtube.com/@HelloFromSpaceAus)), presented by Brad, an Australian space-exploration YouTuber · **Format:** presenter to camera with animated graphics, stock footage and chapter cards · **Published:** 2026-11-11 · **Length:** 35:02 · [Watch on YouTube](https://www.youtube.com/watch?v=NykFQVLuSDA)
> Caption source: on the first two spaced rounds `yt-dlp` hit HTTP 429 and the "confirm you're not a bot" check on the android_vr, ios, mweb, android, tv, tv_simply and web_safari clients, and `youtube-transcript-api` was IP-blocked. On the third round (about 9:37 AM CT) the **android** client returned YouTube's English auto-generated captions (`en-orig`, about 4,650 words). Quotes come from those captions and may contain small transcription errors; names are corrected where clear (the captions give "Frederique Moran and Camille Baloffy" for Frédéric Marin and Camille Beluffi, and "SCE to Ox" for "SCE to Aux"). Screenshots are from the 360p video. Section headings follow the creator's on-screen chapter cards. The video was added from the user's "Summarize" playlist.
## Outline
### Introduction: the "space is big" problem (0:00 to 4:10)
- [0:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=0s) Our galaxy holds an estimated 100 to 400 billion stars, most with planets, some in the habitable zone where liquid water could exist. Reaching them means crossing "many trillions of kilometers".
- [0:56](https://www.youtube.com/watch?v=NykFQVLuSDA&t=56s) Earlier videos on the channel looked at fast options (nuclear pulse propulsion, antimatter drives) that sit "somewhere between technically conceivable and good luck with that". The alternative: "What if we simply stopped trying to get there quickly?"
- [1:35](https://www.youtube.com/watch?v=NykFQVLuSDA&t=95s) The generation ship: people board knowing they, their children and probably their grandchildren will never arrive; generations are born, live and die aboard until descendants reach another world.
- [2:25](https://www.youtube.com/watch?v=NykFQVLuSDA&t=145s) It sidesteps the propulsion problem only by swapping in others: keeping the civilisation, the people and the machine alive through "a 150 year plus journey". He promises to give the idea "every reasonable advantage", and flags that 150 years is already unrealistically short (explained later).
![[images/Why-Generation-Ships-Cant-Work-t0020.webp]]
*Brad introduces the generation ship idea*
### Chapter 1: A civilisation with no exit (4:10 to 8:01)
- [4:10](https://www.youtube.com/watch?v=NykFQVLuSDA&t=250s) The first generation is the easy one: it can be chosen, psychologically screened, trained together for years and picked for complementary skills.
- [4:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=270s) NASA already treats isolation, confinement, interpersonal conflict, autonomy and differing values as serious behavioural risks on missions of months or years; a generation ship stretches them across lifetimes.
- [5:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=300s) You cannot select descendants: "Every child born aboard the ship is another roll of the biological and environmental dice." Normal human variation becomes dangerous in a closed machine that needs everyone's cooperation, and societies are rarely static for centuries.
- [5:55](https://www.youtube.com/watch?v=NykFQVLuSDA&t=355s) On Earth you can leave a system you dislike; aboard the ship "there is nowhere else to go". The founders volunteered, "but their children didn't". Someone born 50 years in is "too late to get a choice, but too early to reap the rewards".
- [6:55](https://www.youtube.com/watch?v=NykFQVLuSDA&t=415s) The harder ethical issue: a generation may refuse to have children. Having children "becomes a piece of mission-critical infrastructure". "Do we pressure them? If that isn't enough, do we legislate reproduction?"
![[images/Why-Generation-Ships-Cant-Work-t0250.webp]]
*Chapter 1 card: "A Civilisation With No Exit"*
### Chapter 2: How many people do we actually need? (8:01 to 12:27)
- [8:05](https://www.youtube.com/watch?v=NykFQVLuSDA&t=485s) A few hundred people sounds intuitive, but there is no agreed minimum.
- [8:35](https://www.youtube.com/watch?v=NykFQVLuSDA&t=515s) Anthropologist Cameron Smith modelled a roughly five-generation, 150-year voyage and estimated a founding population of 14,000 to 44,000, with a cautious figure around 40,000, "remarkably close to the total crew size of an entire Imperial class star destroyer".
- [9:25](https://www.youtube.com/watch?v=NykFQVLuSDA&t=565s) Frédéric Marin and Camille Beluffi's Monte Carlo code HERITAGE (births, deaths, infertility, accidents, pairing, inbreeding) found that a crew of just 98 could stay genetically healthy over a far longer 6,300-year simulated journey.
- [10:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=600s) The catch: the result depends on "adaptive social engineering", a highly controlled reproductive regime. He stresses that the paper describes simulation rules, not enforcement, but translated into shipboard governance it would be "a fairly huge ethical ding". Its value is as a floor figure to plan margin from.
- [11:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=660s) The numbers do not really contradict each other; they answer different versions of the problem, with different assumptions about risk, catastrophes and "how much control the mission is allowed to exert over people's reproductive lives".
- [11:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=690s) Another option, revisited later: a small crew plus a large cryogenic bank of fertilised embryos to add genetic material over time.
![[images/Why-Generation-Ships-Cant-Work-t0483.webp]]
*Chapter 2 card: "How Many People Do We Actually Need?"*
![[images/Why-Generation-Ships-Cant-Work-t0545.webp]]
*"14,000 - 44,000": Cameron Smith's founding population estimate*
![[images/Why-Generation-Ships-Cant-Work-t0560.webp]]
*Star Wars comparison: an Imperial-class Star Destroyer*
![[images/Why-Generation-Ships-Cant-Work-t0600.webp]]
*On-screen source: Marin & Beluffi, JBIS 71 (2018)*
### Chapter 3: Procreation in space, is this even possible? (12:27 to 17:29)
- [12:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=750s) Adults survive months in microgravity, but a generation ship needs the whole life cycle away from Earth: conception, pregnancy, birth, childhood, puberty and reproduction, generation after generation.
- [13:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=780s) NASA still treats mammalian reproduction and development in altered gravity, even sperm function in microgravity, as an open research problem. "So far as anyone is willing to openly admit, no humans have engaged in space intercourse."
- [13:55](https://www.youtube.com/watch?v=NykFQVLuSDA&t=835s) Without countermeasures astronauts lose weight-bearing bone at about 1 to 1.5% per month; he points to his terraforming Mars video for detail on microgravity and radiation.
- [14:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=870s) The key difference: an astronaut's skeleton, heart and vestibular system developed in 1 g. Nobody knows the minimum gravity needed for normal development, "but we do know it's definitely significantly more than zero", so artificial gravity is essential.
- [15:15](https://www.youtube.com/watch?v=NykFQVLuSDA&t=915s) Spin gravity trade-off from NASA studies: at about 4 rpm a 56 m radius gives 1 g, but Coriolis effects cause tilting and tumbling sensations, vertigo and nausea ("If you've ever been on a gravitron at an amusement park..."). At 1 rpm those effects largely go away, but the radius jumps to 895 m.
- [16:50](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1010s) That means an enormous pressure vessel under permanent rotational stress, vibration, docking loads, impacts and thermal cycling for 150 years: "Material science simply doesn't stop applying because we've left Earth."
![[images/Why-Generation-Ships-Cant-Work-t0750.webp]]
*Chapter 3 card: "Procreation in Space - Is This Even Possible?"*
![[images/Why-Generation-Ships-Cant-Work-t0960.webp]]
*Spin gravity at 4 rpm: a 56 m radius for 1 g*
![[images/Why-Generation-Ships-Cant-Work-t1010.webp]]
*At 1 rpm the radius grows to 895 m*
### Chapter 4: The big ship that needs to be a tiny Earth (17:29 to 20:08)
- [17:35](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1055s) Oxygen, water, food, temperature control and waste removal for 150 years: "we aren't bringing 150 years of bottled water", so the ship needs closed-loop life support.
- [18:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1080s) The ISS life-support system has reached about 98% water recovery, including from humidity, sweat and urine, which NASA sees as a key target beyond low Earth orbit.
- [18:25](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1105s) A generation ship must "carry the ability to continuously recreate supplies": CO2 back to oxygen, waste back to nutrients, plants with controlled light, water and minerals.
- [18:55](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1135s) A HERITAGE-team study estimated about 0.45 km2 of artificial farmland (aeroponics plus conventional) to feed about 500 people a balanced omnivorous diet, roughly 60 full-size football pitches.
- [19:25](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1165s) Ecosystems are not machines: populations fluctuate, microbes evolve, crops fail. Earth absorbs those swings because it is enormous; the ship "has very little margin".
![[images/Why-Generation-Ships-Cant-Work-t1051.webp]]
*Chapter 4 card: "The Big Ship That Needs To Be A Tiny Earth"*
![[images/Why-Generation-Ships-Cant-Work-t1085.webp]]
*The International Space Station, whose life support recycles about 98% of its water*
![[images/Why-Generation-Ships-Cant-Work-t1125.webp]]
*Stock footage of an indoor growing lab*
### Chapter 5: A machine that can never die (20:08 to 24:26)
- [19:55](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1195s) "Every pump circulating its water, every fan moving its air, every light growing its crops eventually breaks." Imagine someone in 1876 building a machine that people's lives depend on in 2026; the ship asks that of nearly everything.
- [20:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1230s) Bearings wear, seals perish, lubricants degrade, insulation ages, pipes corrode, structures accumulate thermal-cycling fatigue, and impacts or radiation events add damage.
- [21:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1260s) Spare parts run out, so the ship must make things (machining, metalwork, polymers, glass, ceramics, electronics, chemistry), and recycling is never 100%. "How do you repair the machine that makes the replacement parts for the machine that repairs the ship?" It needs "something approaching an independent industrial economy".
- [21:55](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1315s) Apollo 12: 36 seconds after launch the Saturn V was hit by lightning, then again; flight controller John Aaron recognised the signal-conditioning fault and called "try SCE to Aux", Alan Bean found the switch and the mission went on to land on the Moon.
- [22:35](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1355s) The same problem near Proxima Centauri b, 4.24 light years away: telemetry takes over four years to reach Earth and the answer another four and a quarter, so advice arrives about eight and a half years after the failure. "Hopefully, it wasn't urgent."
- [23:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1410s) Hence huge onboard libraries and autonomous diagnostics: beyond a few light years "there is no mission control in the conventional sense"; Earth can advise slowly but cannot rescue you or send a part.
![[images/Why-Generation-Ships-Cant-Work-t1210.webp]]
*Chapter 5 card: "A Machine That Can Never Die"*
![[images/Why-Generation-Ships-Cant-Work-t1335.webp]]
*Apollo 12 launch footage, before the lightning strikes*
### Chapter 6: OK... just make the ship smaller? (24:26 to 30:33)
- [24:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1470s) Send 20 people plus a cryogenic vault of hundreds or thousands of genetically diverse embryos: the living population stays tiny while the genetic one is huge, with far less habitat, food, life support, shielding and mass, so more performance from the same propulsion.
- [25:25](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1525s) "We've traded scale for fragility." Each crew member is 5% of the population, so refusals, outbreaks, accidents, cancer or fertility problems (especially losing female crew needed for births) bite hard. And "losing 5% of the crew could mean losing 50% of the people who understand some specialized system".
- [26:55](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1615s) "The real kicker": cumulative probability. A 1% annual risk becomes 78% likely over 150 years, 86% over 200, 92% over 250 and 99.3% over 500. That is why he granted the ship an "absurdly generous" 0.028c (2.8% of light speed) at the start.
- [28:25](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1705s) With 100 independent catastrophic failure modes each at 0.1% a year, at least one is effectively certain over 150 years; even at 0.01% a year each, the chance is still 78%. Real ships have far more than 100 failure modes, failures can cascade ("a classic Swiss cheese model failure"), and smaller crews leave less margin to absorb them.
![[images/Why-Generation-Ships-Cant-Work-t1468.webp]]
*Chapter 6 card: "OK... Just Make The Ship Smaller?"*
![[images/Why-Generation-Ships-Cant-Work-t1730.webp]]
*Stock render of a ship in orbit amid debris*
![[images/Why-Generation-Ships-Cant-Work-t1765.webp]]
*On-screen graphic: "Cumulative probability = 99.9%"*
### Wrap-up: Where do we go from here? (30:33 to 35:02)
- [30:40](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1840s) The fast option has two obvious problems: propulsion to a large fraction of light speed, and shielding, since a milligram of dust at 0.5c releases "over 3 tons of TNT" of energy.
- [31:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1890s) Those big, visible problems play "a bit of a trick on our perception": the generation ship has no single equally daunting obstacle, so it feels like just a bigger version of what we have already done. "The devil is in the details", and the idea dies "by a thousand cuts".
- [32:55](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1975s) The fatal flaw is its defining feature, slowness: more time means more wear, accidents, illness and social breakdown, until rare events cross "from possibility into probability".
- [33:25](https://www.youtube.com/watch?v=NykFQVLuSDA&t=2005s) The choice is between speed (a slim chance at an attempt) and a slow ship (a better chance at an attempt, but "a virtually guaranteed probability of... complete catastrophe midjourney"). "The easiest option is probably also the hardest. We need to figure out how to go faster."
- [34:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=2040s) Channel note: after time off he aims to post about once a month.
![[images/Why-Generation-Ships-Cant-Work-t1835.webp]]
*Wrap-up card: "Where Do We Go From Here?"*
## Key ideas
- The generation ship swaps one hard problem (speed) for many quieter ones: social stability, consent, reproduction, gravity, ecology, maintenance and isolation from help.
- Minimum-crew estimates (98 vs 14,000 to 44,000) differ mainly in how much risk is tolerated and how much control over reproduction is assumed.
- Artificial gravity is mandatory for children growing up aboard, and comfortable spin rates force very large structures.
- The ship has to be a self-sustaining ecosystem and an industrial economy, not just a vehicle.
- Shrinking the crew cuts mass but concentrates risk.
- Time is the enemy: cumulative probability turns small yearly risks into near-certainties over a century or more, so the slow ship's main advantage is also its fatal flaw.
## Claims check
| Claim | Verdict | Notes |
|---|---|---|
| 100 to 400 billion stars in the Milky Way ([0:05](https://www.youtube.com/watch?v=NykFQVLuSDA&t=5s)) | Accurate | The standard range of estimates. |
| Cameron Smith estimated 14,000 to 44,000 founders for a ~5-generation, 150-year voyage, with a cautious figure around 40,000 ([8:35](https://www.youtube.com/watch?v=NykFQVLuSDA&t=515s)) | Accurate | Smith (Portland State University), *Acta Astronautica* 2014, "Estimation of a genetically viable population for multigenerational interstellar voyaging". |
| ~40,000 is close to the crew of an Imperial-class Star Destroyer ([9:05](https://www.youtube.com/watch?v=NykFQVLuSDA&t=545s)) | Roughly accurate | Star Wars reference books give about 37,000 crew plus about 9,700 troops. |
| Marin and Beluffi's HERITAGE simulation found 98 founders enough for a 6,300-year journey, using "adaptive social engineering" ([9:25](https://www.youtube.com/watch?v=NykFQVLuSDA&t=565s)) | Accurate | Frédéric Marin and Camille Beluffi, *Journal of the British Interplanetary Society* 71 (2018), on a trip to Proxima Centauri b; the captions mangle both names. |
| NASA treats isolation, confinement and team conflict as serious behavioural risks ([4:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=270s)) | Accurate | These are named risks in NASA's Human Research Program. |
| NASA still treats reproduction and development in altered gravity as an open research question; no human has conceived in space ([13:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=780s)) | Accurate | No human conception or pregnancy in space has been reported; animal studies give mixed results. |
| Weight-bearing bone loss of about 1 to 1.5% per month without adequate countermeasures ([14:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=840s)) | Accurate | NASA's commonly cited figure. |
| 1 g needs about 56 m radius at 4 rpm and about 895 m at 1 rpm ([15:50](https://www.youtube.com/watch?v=NykFQVLuSDA&t=950s), [16:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=990s)) | Accurate | From a = ω²r: about 55.9 m at 4 rpm and about 895 m at 1 rpm. |
| The ISS life-support system recovers about 98% of water ([18:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1080s)) | Accurate | NASA announced 98% water recovery in June 2023 after adding an improved urine processor. |
| About 0.45 km2 of artificial farmland would feed ~500 people a balanced omnivorous diet, ~60 football pitches ([18:55](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1135s)) | Accurate | Marin, Beluffi-Marin and Fischer's follow-up HERITAGE study (JBIS, 2020). A full-size pitch (105 × 68 m) is about 7,140 m2, so 0.45 km2 is about 63 pitches. |
| Apollo 12 was struck by lightning 36 seconds after launch and again later; John Aaron's "SCE to Aux" call, carried out by Alan Bean, saved the mission ([22:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1320s)) | Accurate | Strikes at about 36.5 and 52 seconds after liftoff on 14 November 1969. |
| Proxima Centauri b is 4.24 light years away, so a question and answer take about 8.5 years ([22:40](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1360s)) | Accurate | Proxima Centauri is about 4.24 light years away. |
| At 0.028c the trip takes about 150 years ([28:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1680s)) | Accurate | 4.24 ÷ 0.028 ≈ 151 years, ignoring acceleration and deceleration. |
| A 1% annual risk gives 78% over 150 years, 86% over 200, 92% over 250, 99.3% over 500 ([27:25](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1645s)) | Accurate | 1 − 0.99ⁿ gives 77.9%, 86.6%, 91.9% and 99.3%. |
| 100 failure modes at 0.1% a year give effectively 100% over 150 years; at 0.01% a year, 78% ([28:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1710s), [29:30](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1770s)) | Accurate | 1 − 0.999¹⁵⁰⁰⁰ ≈ 99.99997%; 1 − 0.9999¹⁵⁰⁰⁰ ≈ 77.7%. The on-screen graphic rounds the first to "99.9%". |
| A 1 mg grain at 0.5c releases energy equal to over 3 tons of TNT ([31:00](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1860s)) | Accurate | Relativistic kinetic energy ≈ 1.4 × 10¹⁰ J, about 3.3 tonnes of TNT. |
## Things mentioned
### Books and texts
- No books are mentioned.
- Studies cited: Cameron Smith's population-genetics estimate for a 150-year voyage ([8:35](https://www.youtube.com/watch?v=NykFQVLuSDA&t=515s); published in *Acta Astronautica*, 2014); Frédéric Marin and Camille Beluffi's HERITAGE crew-size simulation ([9:25](https://www.youtube.com/watch?v=NykFQVLuSDA&t=565s); shown on screen as "Marin & Beluffi, JBIS 71 (2018)"); a later HERITAGE study on space farming ([18:55](https://www.youtube.com/watch?v=NykFQVLuSDA&t=1135s), not named in the video); NASA artificial-gravity and Human Research Program material (not named).
- The channel's own earlier videos: nuclear pulse propulsion and antimatter drives ([0:56](https://www.youtube.com/watch?v=NykFQVLuSDA&t=56s)) and terraforming Mars ([14:15](https://www.youtube.com/watch?v=NykFQVLuSDA&t=855s)).
### People
- Cameron Smith, Frédéric Marin, Camille Beluffi, John Aaron, Alan Bean
### Organisations, missions and things
- NASA, International Space Station (Environmental Control and Life Support System), Apollo 12, Saturn V, Proxima Centauri b, Star Wars (Imperial-class Star Destroyer), the "gravitron" amusement-park ride
### Concepts
- Habitable-zone exoplanets, generation ships, consent of descendants, minimum viable population, Monte Carlo simulation, adaptive social engineering, cryogenic embryo banks, microgravity bone loss, vestibular system, spin gravity and the Coriolis effect, closed-loop life support, aeroponics, self-replicating industry, communication lag, cumulative probability, the Swiss cheese model of failure, relativistic dust impacts
## Biographies
### Brad (Hello From Space)
Australian creator of the Hello From Space YouTube channel (@HelloFromSpaceAus), which covers rockets, spaceflight and space exploration; his channel page introduces him only as "Brad". In this video he says he is returning after time off and aims to post about once a month.
### Others mentioned
- **Cameron Smith**: anthropologist at Portland State University who writes on human evolution and the anthropology of space colonisation, including *Principles of Space Anthropology* (2019).
- **Frédéric Marin**: astrophysicist at the Strasbourg Astronomical Observatory (Université de Strasbourg/CNRS) and lead author of the HERITAGE generation-ship studies.
- **Camille Beluffi**: French particle physicist and scientific consultant, co-author of the HERITAGE studies.
- **John Aaron**: NASA flight controller (electrical, environmental and consumables officer) whose "SCE to Aux" call saved Apollo 12 in 1969; he later played a key part in the Apollo 13 recovery.
- **Alan Bean** (1932 to 2018): Apollo 12 lunar module pilot, the fourth person to walk on the Moon, and later a painter.