I seem to have become CT’s resident moderate techno-optimist. So let me push back a little: here are five things that we’re not going to see between now and 2050.
1) Nobody is going to Mars. Let me refine that a little: nobody is going to Mars and coming back alive. A one-way suicide mission is just barely plausible.
[spoiler: he does get home]
“But Elon Musk says” okay you can stop right there.
“But 25 years is a long time! We did Apollo in just like 8 years!” Mars is harder than the Moon. Much, much harder. Traveling to the Moon took three days. Traveling to Mars will take about two years round trip and will require a considerably larger spacecraft.
NASA has been looking at Mars sample return for a while now — meaning, getting a basket of rocks and soil back from the Martian surface, so we can examine them in labs here on Earth. They quickly saw cost estimates balloon into the billions and backed off. Mars sample return is in suspended animation right now, and it’s certainly not going to happen before 2033 at the earliest. Getting humans to Mars and back alive? would be much, much, much more difficult than that.
It would also require at least some level of in-situ resource utilization on Mars. That means stuff like getting water out of Martian ice, for drinking water and possibly propellant. Which is absolutely possible — I’m confident we’ll do it at some point — but we have barely started to think about this yet, and are at least a decade away from even piloting something to try it on a small scale.
Here’s a fun detail: until a few years ago, we didn’t realize that much of Mars’ surface is soaked in chlorates and perchlorates: basically the stuff you find in household bleach. Turns out the chemistry of Mars’ crust and surface is quite different from Earth’s! Also that any Mars travelers will have to deal with the perchlorates somehow. Is that the last potentially dangerous surprise Mars will have for us? Probably not.
There are a bunch of unsolved technological problems with going to Mars, some of which we’re working on — we just made a modest breakthrough in zero-G electrolysis of water — some of which, la la laaaa, we are not. When I see those problems mostly solved, I’ll start to think we /might/ go to Mars. But I’m not holding my breath.
Note that various stakeholders have a vested interest in talking like we really are going to Mars, any day now! NASA has historically been the worst offender here, because reasons, but there are several others. So anything discussing Mars travel? You want to look hard at who is writing it, and consider their motives.
2) Speaking of space woo, we are not going to see asteroid mining. Do I even have to go into this? Briefly: it’s really hard to reach an asteroid and bring anything back, and oh by the way nobody has yet found anything on an asteroid remotely worth bringing back to Earth.
I’m skeptical whether we’ll see meaningful use of asteroid resources in this century at all, but we certainly won’t see it by 2050.
— BTW, I’m actually a huge fan of space science. I think we should be putting balloons in the atmosphere of Venus, deploying solar sails for trips to Mercury and the asteroids, and sending off another interstellar mission to replace the aging Voyagers. A Pluto orbiter? A Mercury lander? Hell yes. Raise my taxes and drip that stuff right into my veins.
But precisely because I take space exploration seriously, I sharply dislike space woo. Manned trips to Mars are woo.
(Oh, and protip: if anyone starts talking about getting Helium-3 from the Moon, you can promptly discount anything they have to say about pretty much anything. No, don’t thank me. Public service.)
3) Coming down to Earth, we are not going to have commercial fusion power. We probably will have contained, continuous fusion reactions — I’m mildly optimistic on this, and won’t be shocked if it happens in the next 10 years. We might perhaps have a reaction that generates more electrical power than it consumes, though I’m less sure about that one.
But commercial fusion power? Meaning, even remotely cost-competitive with coal / natural gas / hydro / wind / solar? No, that’s not going to happen. In the very unlikely event anyone is reading this in 2050: if there’s a commercial fusion reactor producing electricity whose socket cost is no more than three times that of coal, natgas, wind, solar or hydro, whichever is most expensive, I lose. I don’t expect to lose.
By the way, the world is currently getting about 30% of its total electricity from renewables — wind, solar, hydro, a bit of biomass. That’s up from about 18% around the turn of the century. (It’s up a lot more in absolute terms, because world electricity consumption has more than doubled in the last 25 years.) By 2050 that’s expected to be well over 40%.
Whoops, some techno-optimism slipping in there… anyway, point being it’s not clear how much of an incentive there will actually be to commercialize fusion, because we’ll probably be able to go largely carbon-free without it.
4) There will be no superconductor revolution. Superconductors are amazing, and they let you do a lot of neat stuff. But so far they only work at very low temperatures or very high pressures.
[works great as long as you keep pouring the liquid nitrogen]
So STP (Standard Temperature and Pressure) superconductors are kind of a holy grail of materials science. People have been working on them for decades now. Will we see a STP superconductor in the next decade? I would be surprised but not shocked.
But even if we develop a material that superconducts at standard temperature and pressure, it probably won’t be that big a deal. Science nerds tend to get excited about superconductors for one set of reasons, tech bros for another. But both those groups are kind of excitable, you know? A plausible STP superconductor would be nice, but neither revolutionary nor transformative.
Imagine that starting tomorrow we had a superconducting material with a cost and physical properties similar to zinc. Meaning, soft but not very malleable or ductile; rather brittle; low-ish melting point; not very reactive. (Nothing special about zinc, to be clear. This is a thought experiment.) And let’s further say it’s no more expensive than silver (around a dollar a gram, give or take) and can be produced at scale. Most of the currently plausible recipes for STPSCs involve weird alloys and very complicated recipes, so those are actually pretty optimistic assumptions, but let’s be generous.
Okay then, would it be useful? Sure, all sorts of ways. Would it transform our lives, or indeed any particular industrial sector? No, not at all. Zinc doesn’t work very well for long-distance power lines (not very ductile, too brittle), and anyway it would be too expensive.
We’d have cheaper MRIs, sure. And much cheaper particle accelerators. Maglev trains become more competitive, though they’d still be a niche application. Fiercely strong magnets becomee widely available, which is nice. So, various incremental improvements. But a revolution? Not even close.
“Well what about a superconductor that /is/ malleable and ductile, and not brittle, and easily cast or worked? And also very cheap and easy to produce, not requiring any ingredients that are very rare or difficult to handle? And also very chemically stable and not reactive or flammable or explosive or toxic? What then?” Well I feel there should be a pony in there somewhere, you know? That’s a long list, and there’s no reason whatsoever to think a hypothetical STPSC would tick all those boxes. It almost certainly won’t.
(I know where this stuff comes from, unfortunately. There were science fiction stories in the 1970s where STP superconductors that were as cheap and convenient as plastic wrap were a key plot point. In reality, “cost like silver, properties like zinc” is setting the bar very low. At least zinc is a metal, easy to work and handle, and not particularly flammable, toxic, or radioactive.)
5) There will be no useful new physics. No anti-gravity, telepathy, faster-than-light communication or travel, time-travel, teleportation booths, force fields, manipulation of the strong or weak nuclear forces, or reactionless drives. We’re not going to get energy from the vacuum, or perpetual motion, or glowing blue cubes. 
[well, darn]
More to the point, by 2050 we will not have any plausible prospect of any of these things. Like, in 1940 nuclear fission was new physics, and nuclear energy was a distant dream. But people could (and did) claim with a straight face that we would have commercial atomic power within 20 years — and we did!
But there’s nothing like that with new physics today, and there won’t be in 2050 either.
I’m getting very slightly out over my skis on this one, because in theory new physics could surprise us. But 1) we haven’t had any serious, major new physics for a while now — between 20 and 50 years, depending on your definition; and 2) the new physics that we have had? has been interesting but not particularly useful; and 3) almost all the places we’re currently looking for new physics are places where practical applications are extremely unlikely.
I mean, I personally really want to know what dark matter is, whether gravity can be quantized, whether the Koide formula really means anything, and what the deal is with neutrinos. (Seriously, what is the deal with neutrinos.) But that information almost certainly won’t have any practical use whatsoever.
[seriously, what is the deal with neutrinos]
So while this prediction isn’t absolutely airtight, I’d be comfortable betting money on it.
Coming at it from another direction: we’ve been looking at the universe really hard now, at scales large and small, with increasingly sensitive instruments, for over a century. So if there is new physics? Its effects are very likely to be very weak, or to show up only at very large distances or very high energies. So, potentially very interesting, but not likely to be useful.
6) Airships. Zeppelins and dirigibles, yeah? People have been trying to make airships work for a very long time now. The first prototype airship flew in 1854 — that’s not a typo, it was unmanned and steam-powered — and large airships date back to the 1880s. 
[1930s New York, and yes it must have been a hell of a view]
Unfortunately airships are fragile, labor-intensive, and vulnerable to bad weather. They require a lot of room, a fair amount of specialized infrastructure, and — this is the kicker — they’re actually less efficient than other forms of transportation. They’re much slower than airplanes, yet much more expensive than trains or trucks. And no amount of technological innovation has been able to budge those stubborn facts.
There was a vogue a while back for “airships will be useful in places where there aren’t roads! Like ummm Africa!” Except Africa has roads. They’re not always in great shape, but they exist. The Congo has roads. Labrador has roads. Antarctica has roads. People want to go to a place? Before long, there’s a road.
(At this point someone usually mentions Sergey Brin’s Pathfinder. Billionaire’s toy, and if they have a business plan that makes any sense they’re keeping it well confidential.)
Airships look cool as hell, so people will keep trying. But in over 100 years, they haven’t broken out of a few niche uses. It’s not going to happen in the next 25 years. Honestly, I don’t think it will happen ever.
Well then! Let’s check back in 2050, and we can see if I was right.
{ 46 comments… read them below or add one }
NomadUK 09.09.25 at 9:07 pm
Such a Debbie Downer! But probably right. I became seriously disillusioned with everything when 2001 came and went and we hadn’t even started on that spinning wheel of a space station, much less Clavius Base.
I know where this stuff comes from, unfortunately. There were science fiction stories in the 1970s where STP superconductors that were as cheap and convenient as plastic wrap were a key plot point
Larry Niven was probably the biggest spewer of superconductor magic of the time. He was tossing that stuff around like it was — well, yeah, plastic wrap.
The future ain’t what it used to be, and, what’s more, it never was.
bruce.desertrat 09.09.25 at 9:45 pm
Kelly and Zach Weinersmith have recently written a book ‘A City on Mars’ examining just what has to happen for us to colonize Mars, and, minor spoiler, all the subtitle bits with question marks follow Betteridges Law…
https://smbc-store.myshopify.com/products/a-city-on-mars
Cranky Observer 09.09.25 at 10:20 pm
A sixth technological event we shall not see in our lifetime: Doug Muir’s eminently reasonable technology summary published on ArsTechnica [1].
[1] I’ll give a link on Hacker News a non-zero probability, say 0.001%
Alex SL 09.09.25 at 10:26 pm
I don’t know anything about superconductors or quantum computing, but as biologist who also was really into reading about astronomy as a child, the whole Mars boondoggle really annoys me. People have no understanding of the distances and radiation involved in space. They do not understand that Mars has lost its atmosphere and will never again be able to hold one. Colonising Antarctica during an ice age is much more feasible. Mars colonisation is a fraud, a scam, complete nonsense, an indication of a lack of general education.
Brett 09.09.25 at 11:32 pm
It’s more like 3-4 years, rather than two. The 600-ish days missions spend all but about 30-40 days in space, which is not only much harder on the crew but also much more dangerous – they’ll take a higher radiation dose over the whole trip and have to deal with fun stuff like flying past Venus in the inner solar system.
Sample Return ballooned in cost because of NASA’s perfectionism*, but also because they imposed a bunch of requirements that wouldn’t really apply to a human mission. You’re not going to have a complicated, entirely automated orbital rendezvous, and making fuel and oxidizer on the surface means you can send back more. I also don’t think they’d bother trying to harvest and electrolyze sub-surface ice – they’ve got some pretty hefty rockets, so you’d just ship the liquid hydrogen out there in heavily insulated tanks and use it for making propellant with the Martian air.
You deal with the perchlorates by cleaning off your spacesuit, and rinsing anything that might be utilized in unfiltered contact with people with water (perchlorates dissolve in water).
Five years from now? Doubtful. Ten years? I think so with enough money.
Agreed on asteroid mining. That might change with much cheaper launch costs, but those perversely also undermine any economic case for asteroid mining.
Unless something like Helion Energy works out (which is itself a longshot), I agree. Worth noting that all the proposed fusion reactors besides Helion boil steam to generate power, which means you’re probably not going to see truly gigantic cost savings with a fusion reactor even if you get something that’s incredibly reliable and produces far more power than it consumes.
STP Superconductors will likely have a pretty low level of current they can carry through them, if prior high-temperature superconductors are any guide. I think that could be a big deal for small electronics and such, but not a revolution.
Agreed on this. Large airships basically have no niche – if time is more valuable than money, you send something by plane. If money is more valuable than time, you send it by train or boat (or probably autonomous trucks in the near-ish future).
Brett 09.09.25 at 11:37 pm
Sorry, I should have actually added that asterisk note.
NASA puts a huge amount of effort into making each spacecraft as incredibly reliable as possible. This means they often last quite long, but it also makes them much more expensive – I remember talking with Cassini Engineer Frank Crary over NASA Watch about how there was a huge increase in price going from a 20% rate of failure to a 5% or less rate.
Since they’re more expensive, they fly much less often. That means the pressure to stuff them with as many instruments as possible increases dramatically, because you’re talking about potentially the only time in space scientists’ careers where they’ll get an instrument on a spacecraft. Stuffing it, of course, then makes the reliability much harder, which increases the cost, makes for fewer spacecraft, etc.
What we need is more standardization in robotic spacecraft plus more frequency at the expense of a higher failure rate per flight.
John Q 09.10.25 at 2:05 am
Unsurprisingly, I agree on all of these, especially Mars
https://crookedtimber.org/2018/12/13/no-planet-but-this-one/
I don’t recall superconductors playing a big role in the SF I read. I was enthused about the brief period when temperature records were being broken all the time, but I never expected much more than cheap electricity transmission, so I was only mildly sad when it didn’t pan out.
I’ve been more a sceptic about technology (and was burned by my enthusiasm for self-driving cars), so i should probably write something positive as a companion piece to Doug’s
Alan White 09.10.25 at 2:12 am
As usual DM is the voice of reason. Will he turn his to the end of democracy in the US? I’m a lifelong consumer of news, but all of it now gives me indigestion.
Jim Harrison 09.10.25 at 2:45 am
Musk’s Mars obsession is a modern version of the Gilgamesh myth. Unsurprisingly, he and the other tech bros dream of immortality, too. I guess the private island bit isn’t enough.
Gareth Wilson 09.10.25 at 3:26 am
There is an example of finding new physics that supports your prediction. All nuclear technology depends on the strong nuclear force. When quarks were discovered in the 1970s, a complete quantum description of the strong force was finally possible. The practical application were… absolutely nothing. Nuclear bombs, nuclear fission power, even attempts at controlled nuclear fusion, they all use essentially 1940s physics. So maybe we get the same anticlimax if we make another fundamental discovery.
Ken_L 09.10.25 at 4:16 am
I for one would happily kick a few bucks into a fund to send Elon on a one-way suicide mission to Mars.
Doug Muir 09.10.25 at 7:07 am
@5 — “You deal with the perchlorates by cleaning off your spacesuit, and rinsing anything that might be utilized in unfiltered contact with people with water (perchlorates dissolve in water).”
Where are you getting the water?
We think Mars has subterranean aquifers, but they’re probably really deep. There are places where there’s buried ice near the surface, but outside the polar ice caps there’s noplace where there’s ice /at/ the surface. Landing on or near the northern ice cap is one set of problems. Bringing a bulldozer and excavation / drilling equipment is a different set of problems. You might be able to cook water out of hydrated minerals, but that’s yet a third set of problems.
You need energy for all that, and more energy to distill out your pure water. Which means unless you brought a nuclear reactor along you have to first set up a lot of solar panels. Okay, now you have a lot of infrastructure outdoors, So your panels, cables, bulldozer, drill, and everything else better be long-term resistant to perchlorates (as well as dust, temperature swings and radiation obviously). Oh, and if you’re going to rinse the perchlorates off your spacesuit, then your spacesuit must be able to get wet without anything important shorting out, and it can’t have any wrinkles or crannies where perchlorate dust might hide.
Note that using ISRU water, while a great idea, dramatically constrains your choice of landing sites. Also, you probably want to drop a probe on your landing site first, to make very sure it has the resources you’ll need. Which is fine, but it means you’ll need at least one Hohmann cycle of lead time.
TBC, all of these are solvable problems and NASA is putting thought into them. Like, the spacesuit issue? Gets easier if your spacesuit never comes indoors — you attach it to the outside of the hab and crawl in and out through a flap in the back. Awkward, but it eliminates a bunch of other problems. So they’re testing suit designs for that.
But they’re hard problems, and we’ve barely started to scratch at them.
Doug M.
Richard Melvin 09.10.25 at 2:43 pm
The one I might take issue with is asteroid mining. There are no fundamental technological barriers, has been done on a small scale, and is made a lot easier by other recently-developed and developing technologies like solar power and robotics.
The problem is economics. In principle the returns are there; a hundred tons of gold is worth about five billion dollars. The problem is that if you invest in that, you risk success, crashing the price of gold to that of aluminium. If your system works, it is very unlikely you could maintain enough of a monopoly or technical lead to pay back your investment. So the lives of 7 buillion people are made very slightly better by cheaper gold, and you go bankrupt.
This is why most smart investors prefer to invest in scams . These can be prolonged indefintiely without ever risking delivering useful value to anyone.
Kenny Easwaran 09.10.25 at 3:17 pm
I count 6 things on this list of 5!
From least to most likely I would probably rank them as:
Person on Mars, Economically relevant new physics, Superconductor revolution, Asteroid mining, Commercial fusion, Airships.
I don’t think commercial fusion or airships are actually likely, but there’s actually motion towards these things in a way that wouldn’t make them completely out of the question in the next 25 years.
(For reference on superconductors, the lithium ion battery Nobel prize work was between 1980 and 1985, and within 25 years we had things like the iPod and the Segway, and smartphones, e-bikes and commercially successful electric cars not too much later.)
Lee A. Arnold 09.10.25 at 3:50 pm
Solar power IS from a fusion reactor.
Renewables slogan: “Nuclear energy, located at the proper distance from your backyard!”
Hidari 09.10.25 at 5:04 pm
Excellent list, especially the stuff about Mars. One I would add is that I doubt that by 2050 we will have a genuinely conscious machine/computer in the ‘common or garden’ sense of the word conscious: i.e. not a ChatGPT like ‘glorified predictive texter’. Which is not to say that I think that people won’t think we have: according to the Guardian, women are already starting to ‘fall in love’ with their ChatGPT ‘partners’ so people can delude themselves about anything. But I very much doubt that any serious computer scientist will agree with them (i.e. these women).
Please note I’m a materialist: I have no doubts whatsoever that a genuinely conscious machine is a ‘doable project’. I just think that, like a trip to Mars (something else that is eminently doable) that it will be much, much more difficult than some people think nowadays and will therefore take far longer to achieve. Many decades at least. Possibly centuries.
afeman 09.10.25 at 6:53 pm
Whenever somebody talks of the Imminent Revival of Airships, I recall being in my grandparents’ basement in the 70s leafing through grandpa’s decade-old Popular Mechanics magazines with the articles about the Imminent Revival of Airships.
Laban 09.10.25 at 8:23 pm
The trouble with solar energy is that the sun goes down and power is gone. But cheap DC superconducting cables might enable solar electricity transfer from sunny but unpopulated deserts to dark but populated places.
Haven’t you remembered molten salt reactors?
https://world-nuclear.org/information-library/current-and-future-generation/molten-salt-reactors
oldster 09.10.25 at 8:30 pm
“I count 6 things on this list of 5!”
And people say technology isn’t improving!
Thanks, Doug — another great post.
It would be especially good if we could succeed, once and for all, in getting people to stop talking about colonizing Mars. Every breath spent on that pipe-dream is a distraction from the real work of saving our own, irreplaceable planet.
If people want to waste their time on airships? Not much damage done. But the Mars illusion is a genuine time-and-energy trap, esp. for a class of bright, tech-oriented kids who really ought to be working out the next decarbonization technology or the next technology to improve life on Earth!
Alex SL 09.10.25 at 10:18 pm
Hidari,
Agree on the main point: I also don’t see why artificial intelligence should be impossible in principle, but LLMs aren’t it. Two caveats, though.
First, although it is reasonably easy for people arguing in good faith to point at something like an LLM and clarify why it isn’t intelligent, I think we might run into trouble when or if better systems are developed. Do we understand cognition well enough to distinguish when we are being deluded by a much better simulacrum of human cognition than what LLMs can manage from seeing actual intelligence in action? Is there actually a better workable test than the Turing one, which lots of humans are currently obviously failing as they hallucinate that LLMs are sentient?
Second, the strongest argument I have ever seen against electronic intelligence is that some effects may be substrate-dependent, meaning that maybe intelligence like ours is only achievable in a biological brain with all the trade-offs it brings (like susceptibility to infectious diseases and aging). We may not care about AGI like ours that deserves to be called a ‘mind’, though, and instead be okay with an artificial system that is at least very good at abstract reasoning.
dilbert dogbert 09.11.25 at 2:36 am
Looks like Mars is full of rocket fuel!!!
Perchlorates are chemicals containing the highly stable perchlorate ion (ClO4?), used as oxidizers in rocket propellants, explosives, and fireworks, and found naturally in arid environments.
Maybe Muir meant chlorides.
dilbert dogbert 09.11.25 at 3:13 am
Re: Dirigibles!!!
Back in the 1970’s I was part of a NASA study of the modern airship. I did the hybrid airship. My calculations says that if you combine aerodynamic lift with gas lift the gas lift goes to zero. The damn thing wants to be an airplane.
The pure gas lift dirigible has to fly low in the troposphere where all that annoying weather is. To fly at 20,000feet, still in the weather zone, it loses half its lifting capacity. I can see its as a cruise ship viewing African animals from low levels. High enough so a wind gust wont smash it to the ground like happened with the Akron.
It was a fun exercise. I got to meet the old Helium Heads. A Navy photographer who would fly in an airplane to the ship via a hook mechanism. He had lots of great photos. I also met Adm Siberlich who captained the 3W around the Atlantic. Also got to met Adm Rosedahl who saved part of the crew when the Shenandoah came apart. Also met the crazy guy who was wanted to use hybrid airships to cruise the Amazon bringing Christ to the natives. Read the book The Deltoid Pumpkin Seed.
Thomas P 09.11.25 at 6:08 am
Going to Mars: Would probably require a thermal nuclear engine.
Asteroid mining: really only useful if we start constructing large scale structures in space so we don’t have to lift everything from Earth´s deep gravity well.
The less known cousin of room temperature superconductivity: a cheap and efficient thermoelectric element. Being able to generate electricity from any waste heat would be very useful as would producing heat more efficiently than a resistance coil. (for applications too small to use a mechanical heat pump)
Future success of AI will be important. I’m not talking about AGI, just the more limited application of optimizing materials science and engineering solutions far faster than any team of humans. We´ve seen it in areas like protein folding, and applications are likely to expand.
Doug Muir 09.11.25 at 6:15 am
@22, yes exactly. The Deltoid Pumpkin Seed was John McPhee writing about a Really Cool airship design that was going to change everything. That was back in… 1974, was it?
Attempts at reviving the airship come along every decade or two — there have been at least two waves of enthusiasm since McPhee’s book.
The Navy dirigibles that launched scout planes have been forgotten by almost everyone, but yeah they were amazing… briefly.
Doug M.
Doug Muir 09.11.25 at 6:31 am
@21, what? No, I meant chlorates. Totally different thing.
And in the context of doing something on Mars, they’re very unlikely to be useful. The solid rocket fuel you’re thinking of is ammonium perchlorate. The “ammonium” part requires nitrogen fixing from a nitrogen source. And Mars has very little atmospheric nitrogen and (as far as we can tell) only trace levels of nitrogen compounds in its crust.
You’d have to find a nitrogen source and then basically set up an entire modest chemical-industrial base to refine your perchlorate out of the regolith, then (somehow) produce your ammonium perchlorate, plus your powdered aluminum reactant — nobody has found bauxite on Mars either — catalyst, binder, thermite oxidizer, and shell.
“It’s rocket fuel!” It’s the raw material for one component of rocket fuel. Nobody’s going to be using it to launch anything.
Doug M.
J-D 09.11.25 at 6:57 am
The general answer to the question of whether human beings have a sufficiently well developed understanding (of anything, whatever it is we’re talling about) to avoid being fooled is: ‘Only sometimes’.
I don’t know. Maybe there is, and maybe there isn’t. But Turing test or any other test, the test still can’t function better than whoever it is that happens to be operating the test.
I know there’s such a thing as superstition, you know there’s such a thing as superstition, but there’s no absolutely foolproof guarantee against superstition. There can’t be. Likewise, there’s no way to get an absolutely foolproof guarantee that I (or you, or whichever other person you care to name) will not attribute intelligence where in fact it doesn’t exist.
Zamfir 09.11.25 at 7:37 am
Doug says: basically set up an entire modest chemical-industrial base
This is quite doable, the Factorio speedrun record is 44 minutes from landing to first rocket launch
Tm 09.11.25 at 7:53 am
I don’t quite understand why the airship made it on this list (perhaps that’s why the titel only counts 5). It’s a fairly old, well established technology, it’s just not very useful because it has been supplanted by faster, more efficient means of transport. There are still Zepellins flying (https://zeppelinflug.de/de/zeppelin-fluege/rundfluege-friedrichshafen). You can also still use sailships and hot air balloons, they have their niche uses.
engels 09.11.25 at 10:08 am
I don’t quite understand why the airship made it on this list
-doesn’t need a runway
-potentially zero emissions
-amazing views
Doug Muir 09.11.25 at 12:23 pm
@28, because “5 things that attract attention and investment but are woo” would have been a clumsy title.
Airships are minor woo, but they’re also pretty pure woo.
Doug M.
Doug Muir 09.11.25 at 12:30 pm
@23 — “Going to Mars: Would probably require a thermal nuclear engine.”
Not required, but yeah it would definitely help a lot if we weren’t constrained by Hohmann orbits.
People have noodled about using ion engines — low accelerations but constant thrust and great ISP. Still a big lift, but at least it’s a technology that actually exists already.
“Asteroid mining: really only useful if we start constructing large scale structures in space so we don’t have to lift everything from Earth´s deep gravity well.”
Bit of a chicken and egg issue.
“The less known cousin of room temperature superconductivity: a cheap and efficient thermoelectric element.”
True! Unfortunately, we’re not very close to that either.
(Resisting urge to write 1500 words about OTEC.)
Doug M.
J-D 09.11.25 at 12:31 pm
It may be that there is some value in trying to find a way to program an electronic digital computer to do things that humans do in the same way that humans do them; but the value of doing so has to be limited by the fact that if you’re seeking something which does things that humans do in the same way that humans do them, humans themselves are already right there.
David Mitchell 09.11.25 at 8:40 pm
Musk has stated that the reason he wants to start a self sustaining colony on Mars is to have a backup if the Earth is destroyed. That means humans have to be able to reproduce on Mars. What are the odds that humans having evolved under 9.8 m/s2 gravity can reliably reproduce under the 3.7 m/s2 gravity on Mars? I know that there are many other obstacles to building a viable Mars colony, but I’m surprised there isn’t more this one.
Alex SL 09.11.25 at 9:46 pm
J-D at 32,
Yes, that’s what I was implying at the end. And it is a continuing puzzle to me right now why there is such an alleged obsession with AGI, because we don’t need to replicate the general intelligence we already have, we need specialised systems that automate tedious tasks away. Three possibilities:
AGI has become conflated with a self-improving super-AI that will cause the singularity. They are different concepts, but you increasingly see the two used interchangeably.
The AI guys want AGI so that they can replace everybody with digital slave-minds and robots to pay zero salaries, and they have no concept of how an economy would look like with zero consumer purchases but >99% of the population doing starvation riots.
The AI guys have even less of an idea of what they are doing than that and are merely saying AGI as part of a buzzword soup that they have learned will get investments from people who have even less of an idea of what they are doing than the AI guys.
Edward Gregson 09.12.25 at 1:35 am
1.) I think people overstate the degree to which environmental factors prevent a Mars mission, because it’s more rhetorically convenient than trying to sum up how a bunch of logistic difficulties, none of which are prohibitive, combine to make such a mission difficult.
For example the perchlorates thing. They’re clearly not that damaging to space hardware given that probes operated on Mars for decades without being affected by their presence. The danger is the astronauts’ exposure to Mars dust while getting unsuited. You already mentioned suit ports, or you could rinse the suits in the entrance with water. You would probably bring water specifically for that purpose and recycle it through distillation. You probably don’t need a lot given that the perchlorates are a small fraction of the trace suit dust and very soluble in water.
Similarly, the radiation aspect. People way overestimate the degree to which the Earth’s magnetic field shields us from space radiation and underestimate the effect of the atmosphere. Beta and gamma radiation is manageable with conventional shielding techniques, solar particle events are predictable in advance and can have a special shelter in the ship made just for them. Heavy ion cosmic rays are the most concerning since no plausible ship configuration can shield them, but a properly designed Mars mission would see the astronauts get a comparable dose to the long-term Mir and ISS missions from which the astronauts had no measurable health effects. Once you’re on Mars, the atmosphere, thin as it is, dramatically reduces the problem.
Mars is more challenging than the Moon, but we are in a much better place technologically than the Apollo program was. We have far more advanced computers and robotics, and space launch costs about 20 times less, and is expected to decrease further. If SpaceX’s Starship rocket upper stage never manages to reenter correctly and be reused, you still end up with a Falcon 9 the size of a Saturn 5.
It’s mostly just that you’ll never get the federal will to go to Mars that they had during the Apollo program (budgets at 5% of US federal budget!) because the whole thing turned out to be hugely expensive and kind of pointless, and without that kind of cost-be-damned, risk-be-damned race, a program never really attains momentum. If that commitment was there, mid 2030s probably would be doable.
2.) Agreed. This is one of those things that might happen in the future, but far enough out that it’s irrelevant as a concern today.
3.) Careful about the bet – if renewables completely run coal out of business, coal-powered electricity might skyrocket in price.
4.) There might be non-obvious applications of room-temperature superconductors in things like computer chips, but there is a reason why James Cameron switched his Avatar McGuffin from unobtainium to alien whale brain immortality serum.
5.) We haven’t fully exhausted the potential of the previous “new” physics. Quantum computing, quantum sensing, etc.
6.) Airships are death traps, whether filled with hydrogen or not. Even today, when you consider how few of them there are, their accident rate is staggering.
Petter Sjölund 09.12.25 at 2:38 am
Not contradicting this, but it is worth noting that outside of protein folding, machine learning in science has had some disappointing results.
Petter Sjölund 09.12.25 at 2:40 am
Link that disappeared from the above post:
https://www.understandingai.org/p/i-got-fooled-by-ai-for-science-hypeheres
KT2 09.12.25 at 2:41 am
Doug said “You want to look hard at who is writing it, and consider their motives.”
Agreed.
Sean Duffy’s first open presser.
“NASA found intriguing rocks on Mars, so where does that leave Mars Sample Return?
“I have to look at dollars, and I have to look at time, and I have to look at return.”
ERIC BERGER – SEP 11, 2025
…
[mumble blather trump cancel blah ra ra 3 options]
…
“However, based upon discussions with several sources, there appear to be three potential paths that NASA is considering:
[ 1,2,3.. commercial, prize, humans!]
https://arstechnica.com/space/2025/09/nasa-found-intriguing-rocks-on-mars-so-where-does-that-leave-mars-sample-return/
Astounding Sean Duffy is ‘in charge’… read overlord…
Sean Duffy
“Following the incident, he repeated Trump’s assertion that the apparent failure of air traffic controllers to avert the collision was a result of diversity, equity, and inclusion hiring practices.[79]”
…
“On July 9, Trump named Duffy as the acting administrator of NASA, succeeding acting administrator Janet Petro.[97]”
https://en.wikipedia.org/wiki/Sean_Duffy
Note to Sean: sack yourself now.
KT2 09.12.25 at 2:50 am
Next article is another (6/7th zero?) technology which will only protect those encapsulated by Withoit’s Law… Golden Dome. We will see it, but only to protect the overlords and…
“The first 21 satellites in a constellation that could become a cornerstone for the Pentagon’s Golden Dome missile-defense shield successfully launched from California Wednesday aboard a SpaceX Falcon 9 rocket.”
https://arstechnica.com/space/2025/09/pentagon-begins-deploying-new-satellite-network-to-link-sensors-with-shooters/
I, and I presume others Doug, would probably appreciate a post from you re the original and best “technological achievements! (That we won’t see any time soon.)”…. Golden Dome.
As you say “You want to look hard at who is writing it, and consider their motives.”
Ken_L 09.12.25 at 7:07 am
The Energy Secretary of the United States (he studied fusion at university, you know):
“Don’t worry too much about planet-warming emissions, the US Energy Secretary has told the BBC, because within five years AI will have enabled the harnessing of nuclear fusion – the energy that powers the sun and stars.”
https://www.bbc.com/news/articles/cqlz5p314z0o
Tm 09.12.25 at 9:40 am
Quantum Computing isn’t on the List but there’s a good chance that this hype too is without substance. It seems that the alleged factorization successes are based on cheating.
https://www.schneier.com/blog/archives/2025/07/cheating-on-quantum-computing-benchmarks.html
Trader Joe 09.12.25 at 11:12 am
I have no disagreements with the list but I guess I’d pose the question of what is the threshold you’re using to define “new physics?”
I’d suggest that the development of graphene and quantum annealing (both heavily used in the production of chips) would qualify inasmuch as they sort of push the boundaries of man using matter for his own purposes rather than just relying on nature. That said, you may be looking at it from a more fundamental science rather than a say physical engineering standpoint.
Thanks as usual for the thoughtful post.
Doug Muir 09.12.25 at 12:13 pm
@35, “Robots handle the environment just fine” is not a terribly compelling argument. Voyager 1 is holding up amazingly well, but we’re not sending humans to interstellar space any time soon.
That said, yes the environmental issues (that we know of) are all solvable challenges — if you have 15 years and a trillion dollars to play with. I don’t claim for a moment that we can’t go to Mars. We totally could! It’s just fantastically expensive and dangerous, and there’s no plausible actor who would be strongly motivated to put in the necessary time and effort.
The real issue isn’t the environment. It’s 1) mass times delta-V, the tyranny of the rocket equation, and 2) keeping humans alive without resupply or help from Earth for hundreds of days at a time. We’re not even close to being able to do that right now! The ISS is darn interesting, but it is /not/ a working prototype for a Mars mission.
The 2050 cutoff is also relevant here. I could imagine a propulsion breakthrough — nuclear, large scale ion — getting out of the lab in the 2030s, piloting in the 2040s and being ready to push something to Mars by 2060. I don’t /expect/ it, but I wouldn’t rule it out.
Doug M.
Doug Muir 09.12.25 at 12:20 pm
@38, putting Sean Duffy in charge of NASA is a tell: they want spectacle (beat China to the Moon!) but otherwise have no interest whatsoever in what NASA does.
(And /will/ we “beat China” to go back to the Moon? Probably not. This deserves its own blog post, but the teal deer is, the only question is whether the Chinese want it badly enough.)
At least when the Russians decided to set their space program on fire, they got some spectacular corruption out of it.
Doug M.
Cranky Observer 09.12.25 at 2:25 pm
There is the minor issue of the effect on the Earth’s environment resulting from the hundreds to thousands of launches needed to get the tonnage of the Mars expedition into Earth orbit, as well as the gas pollution of the near-Earth space.
Laban 09.12.25 at 5:01 pm
Ken_L 40
Even if hydrogen fusion went live tomorrow with the promise of infinite energy, we then have to think of ways to get rid of all that heat. The fate of the Martian atmosphere should be ever before us.