## The morality of quantum computing

This morning a humanities teacher named Richard Horan, having read my NYT op-ed on quantum supremacy, emailed me the following question about it:

Is this pursuit [of scalable quantum computation] just an arms race? A race to see who can achieve it first? To what end? Will this achievement yield advances in medical science and human quality of life, or will it threaten us even more than we are threatened presently by our technologies? You seem rather sanguine about its possible development and uses. But how close does the hand on that doomsday clock move to midnight once we “can harness an exponential number of amplitudes for computation”?

I thought this question might possibly be of some broader interest, so here’s my response (with some light edits).

Dear Richard,

A radio interviewer asked me a similar question a couple weeks ago—whether there’s an ethical dimension to quantum computing research.  I replied that there’s an ethical dimension to everything that humans do.

A quantum computer is not like a nuclear weapon: it’s not going to directly kill anybody (unless the dilution refrigerator falls on them or something?).  It’s true that a full, scalable QC, if and when it’s achieved, will give a temporary advantage to people who want to break certain cryptographic codes.  The morality of that, of course, could strongly depend on whether the codebreakers are working for the “good guys” (like the Allies during WWII) or the “bad guys” (like, perhaps, Trump or Vladimir Putin or Xi Jinping).

But in any case, there’s already a push to switch to new cryptographic codes that already exist and that we think are quantum-resistant.  An actual scalable QC on the horizon would of course massively accelerate that push.  And once people make the switch, we expect that security on the Internet will be more-or-less back where it started.

Meanwhile, the big upside potential from QCs is that they’ll provide an unprecedented ability to simulate physics and chemistry at the molecular level.  That could at least potentially help with designing new medicines, as well as new batteries and solar cells and carbon capture technologies—all things that the world desperately needs right now.

Also, the theory developed around QC has already led to many new and profound insights about physics and computation.  Some of us regard that as an inherent good, in the same way that art and music and literature are.

Now, one could argue that the climate crisis, or various other crises that our civilization faces, are so desperate that instead of working to build QCs, we should all just abandon our normal work and directly confront the crises, as (for example) Greta Thunberg is doing.  On some days I share that position.  But of course, were the position upheld, it would have implications not just for quantum computing researchers but for almost everyone on earth—including humanities teachers like yourself.

Best,
Scott

### 74 Responses to “The morality of quantum computing”

1. Jon K. Says:

Nice response. I wonder if some of his worries come from not knowing what all the implications of QC might be. I think you pretty much assuage those concerns, although if we do have more technological progress in engineering as a result of QC, I do think our near-term existential risks go up. But it’s just a matter of rate of change; technologies are going to become more powerful whether QC is viable or not, so there’s no point in fighting it. Plus, right now we already have climate risk being a short-to-medium-term existential risk, and we’re going to need some good engineering solutions to help us there, so, yeah… Save us QC! 🙂

2. Eyal Says:

Even as a provocation, it goes way to far to compare a clown like Trump to evil masterminds like Putin or Xi Jinping who – just for starts – are probably personally responsible for the killing and/or incarceration of their own citizens just because of their views.

3. Scott Says:

Eyal #2: Trump has told us openly, over and over, that he’d do the same if he could (e.g., jail his political opponents). It’s only our system of checks and balances, or what’s left of it, that’s prevented him.

4. Eyal Says:

(1) He also says he has infinite wisdom, I presume you take it just as seriously
(2) Are A clown and an evil mastermind blocked only by the “remains of our checks and balances” empirically distinguishable?

5. Allan Jansen Says:

Is it conceivable that a quantum computer could simulate more than its own amount of atoms? The implications of that would be massive but I imagine it’s illogical

6. Eyal Says:

Anyway, when you melodramatically exaggerate like this, nobody will take you seriously – don’t we have a president known for this behavior who pays this price?

7. Scott Says:

Eyal #6: I don’t see how I melodramatically exaggerated anything. I described him as a “bad guy,” which I think is a pretty massive understatement of the truth. And I mentioned him in the same sentence as dictators who I think share all essential elements of his worldview, though they differ in being smarter, savvier, and less constrained by their societies’ institutions.

8. JimV Says:

I read today that a judge found Trump guilty of cheating charities via his Trump Foundation and fined him two million dollars; and as we know he has had refugee children separated from their parents. My estimation says he’s 99% evil. Not smart enough to be a mastermind, though. There might be some evil masterminds among the Republicans who have kept him power for their selfish ends.

Anyway, I just meant to say, wow, what a great reply to the humanities professor. I have a feeling, maybe jumping to a conclusion, that I wouldn’t want to take a course from him.

9. William Hird Says:

There is also a potential “metaphysical aspect” to a large scale quantum computer, namely, it might be a sensitive enough receiver to act as a gateway or communication device to the astral world, or the spirit world ( all those who do not believe in these things please ignore my comment 🙂 )

10. Scott Says:

William Hird #9: Would that be an argument for or against building it? 😉

11. Jacob Says:

I think this is a pretty good response to his question. I am not sure if he is truly curious or trying to prove a point. However at the end of the day it is almost an absurd question. Has the classical computer been good or bad for humanity? Has the plow or wheel? I would say yes.

A quantum computer (unlike a nuclear weapon) is literally just a tool.

12. Joshua Zelinsky Says:

# Allan Jansen ,

If you mean can n qubits somehow simulate n+1 general bits, the short answer is no. For most purposes, n qubits acts closer to n bits than n+1 bits. Holevo’s theorem is one of a few results which are relevant here.

13. Zeb Says:

Well, one of the potential applications of a fully fault-tolerant quantum computer is simulating nanotechnology. Better simulations of nanotechnology could lead to nanotech design and fabrication becoming easy and cheap.

Now imagine that some bored nanoengineer decides to find out how hard it would be to actually create a self-replicating gray goo that eats the world…

Of course, such an effective gray goo (like ice-nine) is unlikely to exist. A more realistic doomsday scenario would involve a bored biologist using a quantum computer to simulate a maximally virulent pathogen (but biologists can already create dangerous pathogens with existing technology).

14. James Gallagher Says:

You were very generous to respond so nicely to someone who has such a negative view of scientific progress.

We could point to the case of lasers, which in the early 1960s were just depicted as evil weapons technology and something that could damage James Bond in a very personal area in the Goldfinger movie (And that depth of reply would perhaps be sufficient in this case)

But I am concerned that people like yourself feel the need to deal with such anti-technology people – who are almost saying that humanity went wrong when scientific progress took off a few centuries ago.

Btw, Has that 14 photon Boson Sampling result been confirmed yet?

15. Posthuman liberalism | Singularity Politics Says:

[…] “Will this achievement yield advances in medical science and human quality of life, or will it… […]

16. Richard Gaylord Says:

“The morality of that, of course, could strongly depend on whether the codebreakers are working for the “good guys” (like the British breaking the Nazi codes during WWII) or the “bad guys” (like, perhaps, Trump or Vladimir Putin or Xi Jinping).” i dislike Trump as much, probably more than the next guy (even when that next guy is an ‘never-Trumper’) but lumping him with Putin or Jinping is unjustified, Detention centers and child separation may be deplorable (though no more so than Japanese internment camps or Native American reservations) but they are not the equivalent of Nazi death camps. we need to distinguish between those individuals we disapprove of. Trump is a truly dispicable human being without any discernible (to me) redeeming qualities but he is not a mass murderer and does not run a gulag (as in China) or seek political control over other countries (such as Tibet, Hong Kong, Ukraine). Mussolini was bad but not in comparision to Hitler or Hirohito. It’s important to distinguish between the bad, the deplorable and the truly evil.

17. Craig Gidney Says:

> Is it conceivable that a quantum computer could simulate more than its own amount of atoms?

That’s definitely impossible. At least, not without some sort of caveat like an exponential slowdown in the simulation time or restrictions on what you can do with the simulated atoms.

If it were possible to simulate more atoms than the computer was made of, without any restrictions, you could arrange those simulated atoms into a larger simulated quantum computer within your smaller quantum computer. Within that larger simulated quantum computer you could then arrange an even larger simulated simulated quantum computer. Repeat as many times as needed to achieve any finite computational power, no matter how large, on a finite sized machine.

This would at the very least contradict the space hierarchy theorem, because with a finite sized classical machine you could simulate the fixed size quantum machine simulating a simulated simulated … simulated quantum machine simulating a classical machine of arbitrary size.

18. Rahul Says:

I interpret the question differently: It’s not the worry of QC being itself used as a weapon to kill.

But, if too many resources go into this cause, of course, it could reduce the money available to (say) fighting malaria. Of course, this is subjective.

Furthermore, I feel it’s the wrong response to counter this with “should everyone only work on climate change?” Basically this seems like a knee jerk reaction; why are you questioning spending on my field?

The other counterproductive response I see often is to point to some X which is an even worse use of resources (subjectively) e.g. “When so many millions are spent on weapons or twitter or whatever why don’t you stop those first before you criticize the spending on QC?”

Let’s take the question for what it is: Are too much resources being spent on QC?

I think it’s a fair question. Even if the precise value judgments may be subjective why not at least look at the data?

e.g. What’s the sum total of US government spending (across all agencies) on Quantum Computing? I’ve no idea. Does any one know?

Then let us compare that to the spending on (say) malaria.

I think that would make for an interesting debate with a factual basis. Maybe the spending on QC is indeed reasonable and justified! Maybe there’s no way to make a call. But let’s look at the data at least?!

19. New top story on Hacker News: The Morality of Quantum Computing – Golden News Says:

[…] The Morality of Quantum Computing 5 by weinzierl | 0 comments on Hacker News. […]

20. TrueHistory Says:

Interesting article, just a ‘note’ on the true facts. Actually it is Poles that broke the Enigma Code, not British. I mean, Turing did contribute but before he could do anything he met Polish people who understood that to break the Enigma Code they had to use ‘mathematics’. I am not going to educate you in full but I would rather check the history as it is, not as one might wish it to be.

21. Rahul Says:

“or the “bad guys” (like, perhaps, Trump or Vladimir Putin or Xi Jinping).”

Which recent President (national leader) do you put in “good guy” category?

How about the likes of Merkl or Modi? Which bin are they in?

22. Árni St. Sigurðsson Says:

Nobody has stated the obvious that Mr. Horan makes an extraordinary claim (devloping QC will move us closer to midnight on the Doomsday Clock) for which he provides no evidence or even an attempt at supporting.

23. Jim Hefferon Says:

Perhaps Mr Horan is thinking something like this: in the 70’s we had computers but they were scarce enough and weak enough that they mostly printed paychecks and managed inventory. A person could think that on balance they were a net good. Today there are many more computers and they are individually much more powerful. A person can worry that this orders of magnitude difference in capability has turned computers to a net evil. For instance, Edward Snowden has told us that because we each have these devices on our desks and in our pockets, the NSA could, and does, listen to us and look at us any time it cares to. We can each think of many other cases that are at the very least troubling.

That is to say, just as the order of magnitude difference between speed of walking, speed of a railroad, and speed of an airplane makes essential differences in societies, could the vastly greater capacity of QC change things essentially?

I think most of us need no convincing about the benefits of better medicines, etc. And the need to understand things scientifically is like the need to do art or create music. But a reasonable person could reasonably worry, it seems to me.

24. Matthijs Blom Says:

While not related, the title does make me curious about morality in complexity theory and/or quantum computing specifically, analogous to morality in mathematics generaly as pointed out by Eugenia Cheng in “Mathematics, morally” (http://eugeniacheng.com/wp-content/uploads/2017/02/cheng-morality.pdf).

25. Scott Says:

James Gallagher #14:

Btw, Has that 14 photon Boson Sampling result been confirmed yet?

If you mean, has it been accepted yet by a peer-reviewed journal, then no, not yet, I don’t think so.

26. Jill Says:

Basically what you’re saying is that US is the bad guy. I don’t think so.

27. Denis Kuperberg Says:

It could be argued that Google achieving quantum supremacy, instead of a publicly funded and open research organization, is a hint that it might not be good news. The GAFA already proved repeatedly that profit is more important to them than building a better society. So while in a better world, quantum supremacy would be good news for everyone, in the present context is is likely to increase inequalities, and only profit those who have the financial means to exploit this technology. Big tech companies invested massive amounts of money to get there, and I suspect they expect return on investment. A parallel can be drawn with deep learning, which is now mainly used for advertisement and other applications that do not necessarily improve society. Granted it can also help scientific research, but this is marginal as an impact on society, compared to what directly reaches everyone on a daily basis. This “only for profit” use is likely to be even worse with quantum computers, since deep learning technologies can be relatively easily copied and made open source for everyone, while QC will probably need extremely expensive and specialized hardware for some time.

28. Peter W. Says:

Yes I agree. He has no idea what he is talking about. He compares groups to individuals and thinks that’s okay. You cannot compare “the good guys” (The Allies) to 1 person (the president). China is not a communist country because Xi is a communist, it’s the whole country, it’s never one person. You should’ve just said China and Russia. So yes in your statement basically you’ve said that the USA is the bad guy. And btw I don’t understand why you have to politicize your perfectly fine article, for comments like mine?

29. Scott Says:

Richard Gaylord #16: I actually agree that “Trump=Hitler” is an invalid moral comparison, and it’s not one that I’ve ever made. On the other hand, I don’t think it’s any exaggeration at all to say that Trump would trample human rights to the same degree as a Xi or a Putin the second he was able to. Again, he’s told us so himself many times.

30. Scott Says:

TrueHistory #20: Thanks for the reminder; changed the sentence to talk about “Allies” (which then also includes, e.g., the American breaking of Japanese codes).

31. Scott Says:

Rahul #21: Obama was good, Merkel is net good, Modi is net bad.

32. Scott Says:

Jill #26 and Peter W #28: It’s precisely because the US is desperately needed as the world’s most powerful “good guy” that its capture by a radical faction that rejects the country’s founding ideals and democratic norms—namely, the Trumpists—was so catastrophic, as if our nation were defeated in war. The only good news is that the defeat need not be permanent.

33. William Hird Says:

@ Scott #10.
“Would that be an argument for or against building it ?”
Well for Peter Shor it would be a great way to become famous twice. Fast forward 100 years and engineers are getting ready to fire up the first fault tolerant million qubit QC to run Shor’s algorithm. They turn the machine on and hear the unmistakable voice of Peter declaring, “what took you so long, I’m on the other side now, there is eternal life.” 🙂

34. Rainer Says:

Scott #31,

a question from a german non-native speaker: What does “net good” mean?

35. Scott Says:

Rainer #34: “more good than bad.”

36. Rahul Says:

Nitpicking, but why is Obama “good” vs Merkel “net good”?

37. JAM Says:

Nonsensical article. The article is another example of derangement that the left degenerated to. Putting Putin and Xiaoping next to Trump, legally elected president that is now being under attack to be removed from office and negate election results. If anyone should be put in the same row as Putin is Trump enemies. No technology is immoral. If anything only use of technology can be immoral. Only complete idiot would block progress of technology on the basis of assumption that it can be potentially used for “immoral” acts. And who is there to judge what is immoral? In the western culture it used to be Catholic Church deciding what is and what isn’t immoral. But that era has passed. Why would I value self appointed “intelligentsia” as some kind of morality arbiters? In fact the left killed the whole idea of morality. We can now “morally” kill millions of babies in the womb. It is “immoral” now to point out obvious sexual deviations. It seems that “immoral” now is more of what current leftist think tank says it is. Not what used to done for the good of the society.

38. Michael N. Says:

Hi Scott,

I found this line particularly interesting: “That could at least potentially help with designing new medicines, as well as new batteries and solar cells and carbon capture technologies—all things that the world desperately needs right now.”

How concerned are you with the amount of investment in quantum computing given this uncertainty? I don’t think anyone can argue against quantum computing research continuing to contribute tremendously to basic science. But if someone handed you a fully error-corrected quantum computer tomorrow, do we know for certain that it would fundamentally change the world (past changing encryption, which as you point out is a mixed bag)?

From a relatively uninformed position, I would heavily bet yes. Certainly, we invest in technologies that have unsure futures all the time. But, assuming one should account for uncertainty when valuing a technology, do you think that the enormous investment being made in quantum computing is a saavy one? Even if it is highly likely to change the world, uncertainty in the face of billions and billions of dollars and the collective intellectual investment of an entire academic community seems terrifying.

39. Scott Says:

Rahul #36: Simply because I haven’t studied Merkel’s tenure or policies enough to feel confident in my judgment. On the other hand, anyone who’s helping to hold the line against the world’s resurgent authoritarians is “net good.”

40. Scott Says:

Michael #38: You’re absolutely right; we can’t say for certain right now that there would be world-changing applications (beyond breaking crypto) even if we had a fully fault-tolerant device. Given the uncertainty, I feel like the current degree of investment—a billion here, a hundred million there—is more-or-less commensurate with the promise. I’d rather that additional funding go to support fundamental science more broadly. If we had something like an exp(n1/3) quantum algorithm for 3SAT, or an application where quantum simulation would definitely, absolutely revolutionize engineering, then tens or hundreds of billions in investment might be justified. If NP⊆BQP, then trillions.

41. Scott Says:

JAM #37 and others: You’re welcome, and indeed strongly encouraged, to continue attacking me for being broadly progressive and militantly anti-Trump. Not only does it not bother me, but it could only possibly help with the many people on Twitter, Reddit, etc. who’ve tried to unperson me for insufficient fealty to their social justice religion. 😀

42. Michael MacDonald Says:

I think one possible motivation for the question that you didn’t specifically address is the popular notion that scalable QC will create a breakthrough in AI research that will lead to an artificial general intelligence. You might have mentioned that as far as we can tell, scalable QC and AI are completely separate problems.

43. Scott Says:

Rahul #18 and Michael MacDonald #42: Both of you, in effect, wanted me to answer a question rather different from the one that Richard Horan actually asked! But OK:

Regarding funding for QC, see my comment #40.

Regarding quantum computing and AI: well, “it’s complicated.” 🙂 There’s little doubt that, if we had a full scalable QC, then the quadratic speedup from Grover’s algorithm would help at least somewhat with most of the bread-and-butter tasks of AI. But as for whether we can get larger quantum speedups—particularly exponential ones—for any AI problems that matter in practice, I think the best one can say is that it remains an active research topic for very good reasons; and that alas, many of the exponential speedups that were claimed (or hoped for) in the past have not survived scrutiny.

44. Raoul Ohio Says:

I will buy that Vladimir Putin and Xi Jinping have done more evil than Trump. There are two main reasons for this:

1. The US system of checks and balances is more or less working to restrain Trump.

2. Putin and Xi are smart, clever, and ruthless. Trump is dumb as a rock, an egotistical buffoon, and “wannabe” ruthless.

45. Scott Says:

Raoul #44: Correct. (But what you describe as “the system of checks and balances working,” I’d describe as “the rope has almost entirely frayed, but we’re still, thank God, hanging above the abyss by a thread.”)

46. James Gallagher Says:

Scott, you really are a very negative person at times.

Here’s how to be more positive – think that the the likes of the Trump election (and Brexit referendum result) are welcome checks and balances on the unbridled liberal view that humanity’s problems can be solved by politicians. Of course they can’t, they can only be solved by people themselves in the end, coming to believe in a common way to behave, not because someone has told them to behave like that, or fooled them into thinking they must behave like that (for religious or extreme political reasons)

There is the idea of the benevolent philosopher king, but that way leads dangerously to dictatorship, and in the end, shocking as it may seem, we should be happy that our imperfect democratic systems sometimes throw up not the best outcomes – but at least nothing on the scale of what has previously happened in history (And I really disagree that Trump’s election was equivalent to losing a war – that’s not a sensible view at all)

You know, once people have had a taste of Trump for a few years, it will probably kill that sort of politics for a generation – and he hasn’t been able to cause too much harm, even had some positive impacts on areas of the economy – so pretty good price to pay for just 4 years I suggest – if it leads to a few decades of moderate politics in the USA.

As a UK citizen, I hope our issues with the European Union will be resolved by a moderate trade deal , when all parties realise we all want to get along, let’s just be reasonable about things.

47. Vampyricon Says:

Scott #29,

I would say Xi is much closer to Hitler than your “Trump is not Hitler but will be close to Xi and Putin” (paraphrasing) would imply. Uyghur Muslims in Xinjiang are being put in concentration camps and according to some reports, sterilized by injections. While he isn’t like Hitler in that he isn’t directly killing people, he’s still sterilizing an entire population, and that counts as genocide in the UN’s books.

I’m not familiar enough with Putin to judge whether that’s a fair comparison though.

48. Shmi Says:

Trump, a consummate sociopath and narcissist, given free rein, would likely be more like Gaddafi than like Putin 🙂 Still, so far he has not been nearly as evil as, say, Bush W, and even fired Bolton when pushed by him to start a war with Iran.

49. Scott Says:

James Gallagher #46:

Scott, you really are a very negative person at times.

If you’ll permit a tangent, I had an insight about that the other day.

Maybe you can classify people according to the age at which they realized that the universe is not fundamentally reasonable and just.

Some people were never afflicted by the just-world fallacy, or they snapped out of it as toddlers. These people can ironically have a very positive attitude: they see the injustice and idiocy all around them but they didn’t expect anything better, so it’s fine.

Other people still believe in a basically just universe—or they intellectually accepted the falsehood of that view as teenagers or young adults, but they still believe it emotionally. These people can also be quite happy, since even when they see what looks like injustice or horribleness, they figure that it must be for the best, eventually.

But I lost the belief in a basically just world traumatically, around the age of 7 or 8—the age at which I first experienced the depth of cruelty that other kids could be capable of (while experiencing no negative repercussions for it of any kind), and also the age at which I first understood the Holocaust. As a result, I still experience injustice like a massive shock, a personal affront and wound, and a breach of the contract that I thought I’d signed when I entered the world.

As one instance, the fact that a vindictive charlatan like Trump can not only experience zero negative consequences for his behavior in his entire life, but can actually rise to the Presidency, and start making decisions purely out of spite that imperil the entire world—if you like, you can take the fact that I experience that like a spear to the stomach as a byproduct of my deep-seated American patriotism. Even after George W. Bush, I had still felt in my heart of hearts that the US was fundamentally good—incapable of electing a President who boasted of his own cruelty, and lied in every utterance as a way to prove his dominance. I had imagined that there was an invisible barrier that kept this nation hovering slightly above the abyss that I learned about at the age of 7 or 8—indeed, that despite the Holocaust and all the rest, the continued existence and progress of the United States, more-or-less as envisioned by Washington and Jefferson and Franklin (and having rid itself of the original sin of slavery), was perhaps evidence for some weak form of the just-world hypothesis. Trump’s rise to power permanently shattered that.

50. asdf Says:

I found myself rabbitholing a couple nights ago about a topic called adelic quantum mechanics that I’d previously never heard of. My blurry interpretation is something like this. It’s well known that you can re-develop calculus and analysis over the p-adic number fields instead of the reals. In the resulting distance metric the neighborhood of a point becomes a fractical, tree-like structure instead of a little circle. Doing this over all the primes at the same time gives “adeles”, and re-doing quantum mechanics using adeles instead of reals makes thing combinatorial instead of continuous at the Planck scale, but taking limits gives the usual complex amplitudes at larger scales, plus it unifies some equations found earlier in string theory.

I have no idea whether actual physicists (the ones who see physics as a natural science rather than a branch of math) take this seriously, but apparently as a pure math topic it is considered pretty cool and there is a literature about it. A famous quote by Yuri Manin keeps appearing:

> On the fundamental level our world is neither real, nor p-adic, it is adèlic. For some reasons reflecting the physical nature of our kind of living matter (e.g. the fact that we are built of massive particles, we tend to project the adèlic picture onto the real side. We can equally well spiritually project it upon its non-Archimedian site and calculate most important things arithmetically.

I wonder if this notion appears anywhere in quantum computing, and if it has any predictions. Integers like 2 or 3 or 5 appear all over physics, but I’m surprised that larger ones or combinatorial objects like E8 do (if they really do). The belief that the fine structure constant was exactly 1/137 turned out to be illusory, for example.

I guess this is all pretty woo-woo but I thought it was interesting. There is a web page about p-adic and adelic physics here: http://empslocal.ex.ac.uk/people/staff/mrwatkin/zeta/physics7.htm

51. Allemaraiccire Says:

Since a quantum computer could simulate QCD (right?) then could it be used to better model the nucleus, and give more insight into nuclear fission and nuclear fusion, which could have very good (energy generation) or very bad (weapons) implications? Generally, I wonder about, if a quantum computer could model certain physical or chemical processes (better than a classical computer) could it be nevertheless better to just literally run the actual physical/chemical process itself (i.e. do an experiment) as a model of itself. Also for those actual physical/chemical process that would have been better modeled by a quantum computer rather than a classical computer, could such processes themselves be harnessed as a source of quantum computing power?

52. Scott Says:

Allemaraiccire #51: As someone who doesn’t know nuclear physics, I’ve always found it remarkable just how little humanity’s progress in harnessing fission and fusion depended on any understanding of QCD. And I’m not aware of supercomputer QCD simulations being used today for nuclear energy or nuclear weapons work—my impression was that QCD simulations were pretty much entirely for fundamental physics (like calculating the proton mass from first principles, or understanding the quark soups that you can get in neutron stars or the early universe). So, could using a quantum computer to simulate QCD have any practical relevance for nuclear energy? That’s an excellent question for which I don’t have an informed answer, but I’d be extremely happy for any experts to chime in.

53. Richard Horan Says:

If I may respond to the underlying cause of my concerns regarding QC: In Ray Kurzweil’s book “The Singularity Is Near,” that “singularity” he’s so sanquine about is the point at which computers become self-learning. When that happens, all bets are off. Life as we know it will be beyond our ken of understanding. His vision is in that moment of singularity, “intelligence will grow exponentially and quickly diffuse itself throughout the entire universe.” I don’t know if I want that intelligence snap, crackling, and popping in my neighborhood, especially if it has not earned any right to “know” anything. The universe began with physics, evolved into chemistry, and finally into biology. Biology must then be the manifest destiny of matter. I don’t think we want to fuck with that paradigm. We’re not that smart. If the computers take over, won’t “it” all then revert back to physics even though biological beings created it? And isn’t matter a manifestation of time, intelligently, dilligently, trial-and-errorily transformed? If intelligence, without function–without checks and balances and predator and prey and hunger and famine and sexuality and barrenness is let loose in the environment, what’s the outcome…for us? I don’t envisage a glory there. So, my concern is this underlying idea of self-learning computers, quickly and without intent, growing out of this, because that is what QC is teasing. It really scares the shit out of me, especially the non-biological nature of “intelligence.” Because unlike God, I think the ghost in the machine is more like Trump than Siddhartha Gotama…

54. Scott Says:

Richard #53: There’s a whole community devoted to worrying about exactly the issue you say. It’s called the “rationalist community”; its prominent figures include Eliezer Yudkowsky and Nick Bostrom. (Bostrom wrote a now-classic book about the issue, called “Superintelligence,” in 2014. See also the new book “Human Compatible,” by leading AI researcher Stuart Russell, as well as an excellent popular account called “The AI Does Not Hate You” by Tom Chivers, which I blogged about here.)

The rationalist community agrees with Ray Kurzweil that superhuman AI in our lifetimes is a live possibility, but it takes a nearly-opposite view from him of what it means. Their view is that, unless it’s extremely carefully designed to be “aligned” with human values, a superhuman AI would very likely obliterate life on earth as a prerequisite to pursuing whatever it does value. Many of them even view “AI alignment” as the most urgent problem facing humanity (for whatever it’s worth, Bill Gates, Elon Musk, and the late Stephen Hawking have all strongly supported this cause as well).

You might imagine that the rationalist community would be praised by humanists for its clear realization that technological progress is not an unalloyed good—but you’d be wrong. Today, the in thing to do is to denounce and condemn the rationalists, as nerdy white males trying to distract everyone from more immediate problems, like the evils of sexism and capitalism, and change the focus to a ridiculous adolescent sci-fi fantasy about AI destroying the world.

In any case, as several comments above pointed out, all of these concerns are only slightly related to QC, because QC seems to provide only modest speedups for AI problems.

55. Richard Horan Says:

“You might imagine that the rationalist community would be praised by humanists for its clear realization that technological progress is not unalloyed good—but you’d be wrong. Today, the in thing to do is to denounce and condemn the rationalists, as nerdy white males trying to distract everyone from more immediate problems, like the evils of sexism and capitalism, and change the focus to a ridiculous adolescent sci-fi fantasy about AI destroying the world.” Hmmm, as a boomer, I guess I’m out of the loop on this. I just assumed any truly aware, concerned, astonished, outraged, and disgusted citizen of the planet would have his or her priorities in line. I’m with the rationalists and will follow their thoughts on this one…Thanks.

56. Scott Says:

Richard #55:

Hmmm, as a boomer, I guess I’m out of the loop on this.

OK, boomer. 😀

(I guess I still get to say that? I was born in 1981, which according to Wikipedia makes me among the oldest millennials…)

57. Richard Horan Says:

Unlike my ilk, I’ll eventually figure it out…

58. Simon Devitt Says:

Peter Rohde has been spending an inordinate amount of time thinking about the political, ethical and geo-strategic issues for QC. We had a bit of a chat about it.

https://soundcloud.com/mequanics/meet-the-mequancs-e52-dr-peter-rohde-university-of-technology-sydney

And TEDxNEWTOWN in Sydney has him giving a talk this month (for those in Sydney) that speaks to these issues.

59. fred Says:

TDLR – The ethical concern about QC is somewhere between goat cheese and nuclear weapons.

60. Adam T Says:

FWIW, I read that note from the humanities professor not as an attacking on QC, but rather just a straightforward question on the ethical implications with maybe the implicit forceful suggestion that it should be incumbent on all scientific practitioners to think seriously about the moral and ethical implications of their work. And knowing Scott and seeing this reply I am optimistic that the humanities professor will feel comforted.

Scott, one other point of fact on the Trump=Hitler equation… during the election Trump repeatedly said he’d be willing to torture and/or murder the innocent familial relations of members of ISIS in order to “get tough folks!”

And guess what this was met with by the vast number of people who heard it: silence.

It is still in my mind very revealing and emblematic of his moral compass (or lack thereof) and all those who agree or do not find this wildly awesomely objectionable to the extreme.

62. fred Says:

Not to “defend” him or give him any credit, but the problem with Trump is that you really can’t take what he says at face value. He says things not in the service of truth, but in the service of getting what he wants, or make him look a certain way, or just because he needs to say something to cover his ignorance/confusion. Like a bullying salesman. He doesn’t get that in politics/diplomacy, sometimes words matter as much as actions.

And it goes both ways about him saying something “positive” (praising) or “negative” (threats).
He’s been very loud about using America’s military power to do direct threats of total annihilation (NK, Iran), but I think it’s now clear that he’s not willing to do it at all (clearly less than Bush, and even less so than Obama) – E.g. the Iranians give him plenty of excuses to act.

63. jonathan Says:

Scott,

While I don’t really want to defend Trump, I think that the men and women who happen to work under him cracking codes are on the whole a good deal closer to Allied codebreakers in WW2 than to Putin or Xi’s goons.

I’ll revise my opinion if it turns out he’s been directing them to break into Hunter Biden’s secret Ukrainian email server or something.

64. Allemaraiccire Says:

Scott #52 Any thoughts on my general question (not so on-topic as to the ethics thing, but arises from considering using quantum computer for simulation) about whether, if you have a physical/chemical process that is much better simulated by a quantum rather than classical computer, then could that physical/chemical process itself be harnessed as a source of quantum computing power (perhaps e.g. the electron cloud of a large long molecule)?

65. Scott Says:

Allemaraiccire #64: I mean, yes, a bewildering variety of quantum systems have been studied as possible substrates for quantum computers, following exactly the logic you say. A key sticking point, though, is that to be a good substrate, a system doesn’t merely need to be hard to simulate classically: it also needs to be (relatively) easily controllable, programmable, and measurable.

66. Cambridge Trumpiste Says:

> “good guys” (like the Allies during WWII) or the “bad guys” (like, perhaps, Trump or Vladimir Putin or Xi Jinping)

—————

Virtue signaling, and inserting weasel qualifiers (“perhaps”) for deniability, both look weak, because they are.

It looks rather ridiculous to give Trump zero credit for what he actually has done, such as sign a large increase in QC funding, but full responsibility for things you imagine he would do in some farfetched circumstance. And then to proceed to equate him, on the basis of the imaginary stuff, with a real dictator who is really, personally, factually responsible for large-scale genocide, organ harvesting, totalitarian surveillance, and a state sponsored rape campaign (among other accomplishments).

That kind of rhetoric flies well in Cambridge and Austin. Meanwhile, there is no sign that it bothers you, or has even come to your notice, that equal protection of the law has been suspended for Trump, his attorneys, and associates since 2015. They get “lawfare” and armed FBI raids at dawn, not actual functioning rule of law; their opponents who arrange the dawn raids (McCabe, Comey, Brennan, etc) get book deals and TV gigs. There is literally a show trial going on in the US Congress, and the 2020 election is on track to be decided by which side can indict or arrest leaders of the other. Does any of this in any way trouble you? I ask because all of these irregularities that bear an awfully striking resemblance to Third World coups and banana republic governance have been coming from the Democratic left. But you consistently point to Trump as the sole problem.

67. Allemaraiccire Says:

Scott #65, thanks. It’s interesting to see the juxtaposition of you expertise on quantum computing with you opinions on politics, and while they are in principal unrelated, and there’s not necessarily any contradiction, I do wonder whether evidence and reason would make you completely change these political views. For example, you had often professed a belief that there was some kind of collusion between Trump and Russia, but surely you realize by now, not only that that was always a complete hoax, but that numerous democratic operatives, deep states actors, and US and foreign spies engaged in a huge conspiracy throughout year 2016 to concoct this hoax, in order to undemocratically undermine Trumps’s chances in the 2016 presidential election. Doesn’t this concern you? Do you really want a society where this kind of massive corruption prevails? Also, several prominent Democrats and their family members have long been engaged in massive financial corruption in Ukraine (and China etc etc), as boasted by Biden, and one hard-core Democrat CIA agent, who facilitated this Ukraine corruption in 2016, also leaked confidential classified White House communications in 2017, and then became the so-called impeachment whistleblower in 2019. The same dangerous, powerful, corrupt people have been attacking Trump for years, and it’s because the Democrats are obsessed with power and money at all costs. And they have become very crazy, cult-like, oppressive and totalitarian. They no longer have any ethical principles that are, well, ethical. They brainwash using media and “education” in a way that would make Goebbels envious. Trump is against all this. He’s actually the good guy fighting against this terrifying evil. About half the people can see what’s going on, and half can’t. I wonder what your not seeing. Are you not receiving information? Are you selectively believing whatever fits you current worldview? Are you not actually seeing that current Democrat ideology in extremely insane and dangerous in a way that’s not really a matter of opinion? I really think that if Americans were fully informed, and really thought about what’s really going on, then Trump would get 70%-80% of the vote (including you).

68. Bogdan Says:

Are you going to write a blog post about this paper https://arxiv.org/abs/1911.03748 which proves your conjecture?

69. Andaro Says:

Scott #49: You describe very well a form of pessimism or misanthropy that I see as rational. But I’m not sure if you’ve actually made a Bayesian update on all the implications. For example, if people are generally terrible (which is true) and engage in inherently zero-sum or hostile preferences (which is true), why would the end of civilization be a bad thing? I don’t think climate change is an existential risk as you have claimed, but if it were, why would that be a bad thing? Do you really care if all the nazis and [insert your list of enemy identities] see their systems destroyed? I think most people who want to “save the world” don’t actually care about the consequences, but simply pattern-match to the “hero saves the world” trope that is so common in fiction and is associated with expectations of high status.

I have rationally updated both to the fact that the world is not worth saving and the fact that we don’t have a human rights consensus, which means I don’t have to act as though people had rights (most people don’t act as though I had them, either). At some point, maybe just update to reality?

70. Scott Says:

Bogdan #68: Yes. I wanted time to work through the paper first though.

71. Scott Says:

Andaro #69: If by “optimism” you just mean the belief that the world is worth saving, then I’m 100% an optimist. I’m a pessimist only in doubting the likelihood that it will be saved.

Yes, there’s indescribable evil in the world—but even if you supposed (as I don’t) that everyone complicit in the evil deserves to be consigned to the flames, there would still remain the obvious point that not everyone has had an equal hand in it. Most of humanity strikes me as morally neutral at worst, and a sizable minority strikes me as actively good. And I’d much rather that good people continue to have to fight bad people until the sun goes cold, than (a live possibility) that both of them go extinct in this century.

72. Andaro Says:

Scott #71: Interesting. I have the exact opposite intuition: Civilization causes negative moral value, so extinction would be positive, but I see it as unlikely. It seems you believe the inverse: Civilization causes positive moral value, extinction would be negative, but it is likely. Hm.

I think the biggest factor for me to update was the realization that a very large number of people really aren’t against torturing people and in fact consider eternal torture moral (e.g. because someone deserves it, which can have all kinds of different reasons). There’s really nothing that makes up for that on the positive side, e.g. the small number of people who would provide me with eternal heaven/pleasure doesn’t make up for all the people who would see me burn forever. I mostly just updated that this is what they also deserve, and consider all of us lucky that none of us have the physical ability to implement it (at least not eternally). I’m no longer interested in preventing others from being victimized, so it’s not a negative-utilitarian statement, but I also don’t see why I should want to bear any costs to prevent extinction, e.g. costly GHG reductions.

73. Scott Says:

Allemaraiccire #67: While I don’t have the energy to engage with everything in your comment, let me make one general point. I find most of Trump’s policies (the travel ban, the family separations, the pointless trade wars, abandoning the Kurds, pulling out of the Paris agreement, neutering the EPA..) to be hateful to my values. But even if I didn’t, what I find even more hateful is that Trump explicitly rejects the idea of using reason to figure out right answers. He’s repeatedly fired advisors who disagreed with him until he got yes-men, and boasts of knowing the right answers by following his “gut.” He taunts his opponents like a 4-year-old, or calls for them to be thrown into prison, rather than answering their arguments (which in his worldview would be a sign of weakness).

Thus, there’s something fundamentally strange about the concept of trying to convince someone using reason that they should support Trump. It’s a lot like searching for a proof in ZFC that ZFC is inconsistent. You have almost no chance of succeeding … but if you ever did, the entire project of reason would have collapsed on itself and become self-refuting.

74. Mario J Kidder Says:

Yeah, totally agree, quantum computing isn’t a nuclear weapon but it could change the whole world to make it a better place.

Comment Policy: All comments are placed in moderation and reviewed prior to appearing. Comments can be left in moderation for any reason, but in particular, for ad-hominem attacks, hatred of groups of people, or snide and patronizing tone. Also: comments that link to a paper or article and, in effect, challenge me to respond to it are at severe risk of being left in moderation, as such comments place demands on my time that I can no longer meet. You'll have a much better chance of a response from me if you formulate your own argument here, rather than outsourcing the job to someone else. I sometimes accidentally miss perfectly reasonable comments in the moderation queue, or they get caught in the spam filter. If you feel this may have been the case with your comment, shoot me an email.

You can now use rich HTML in comments! You can also use basic TeX, by enclosing it within  for displayed equations or  for inline equations.

Note added November 11, 2020 / revised November 16, 2020: No longer having time for this, I'll no longer be publishing comments supportive of the ongoing coup d'état against the President-Elect of the United States, except in the unlikely case that they contain an argument I hadn't seen before.