At first I thought she was biased because she was was so skeptical about almost every point during the discussion, but her logic and arguments were all so solid and well thought out, they were impossible to ignore.
I checked out her channel, and she was the first YouTuber that I found who actually explained the syntax of Quantum Mechanics equations and what Kets are. https://youtu.be/ctXDXABJRtg
More detailed content would be greatly welcomed from her.
This “Youtube language” is something a lot of content creators suffer from, unfortunately.
The content of the videos is what matters. The advertising is annoying, but necessary.
https://www.youtube.com/watch?v=S2xHZPH5Sng
The tl;dw is that the most effective titles come right up against the edge of what an average target audience member would consider unacceptable on the axes of withheld information (to create a curiosity gap) and misleading sensationalism (to maximise the perceived importance of the topic); and you have to play the game of having effective titles because otherwise no one clicks on your video and it stops being recommended to people who would find it valuable.
I understand why he did it, but wish he didn't. If I was a leader at YouTube I'd be pretty concerned about this - long term, it drives your platform in the direction of clickbait. Yeah there's benefits to that, but could also turn into a competitive weakness for YouTube. I think they should improve the recommender to make this unnecessary.
What I'm saying is that we can do better than 50 year olds making child-like expressions with clickbaity titles.
But it's extremely distasteful and I'm not willing to lower my standards because of this stupidity.
EDITED to add that what attaches itself to such sentiment (or stokes it), are cynical political actors happy to gain power irrespective of the cost to society.
Don't blame the algorithm. Blame people themselves. It's not YouTube's fault that people find those kinds of titles more interesting.
Of course, that would probably reduce engagement...
Your real complaint is that people like stupid videos, but that's not YouTube's problem or fault.
From the search results I get, it appears that they abandoned that effort long ago.
What she primarily is, is someone who insists on experimental data as an anchor for physical theories. She's not totally against theory that isn't backed by experiment (despite her last paragraph or two), but given multiple competing theories that cannot be experimentally verified, she is unwilling to accept any one of them as "the truth". I don't think she's wrong in that.
https://youtu.be/fxiHM11w-rk (about 2 minutes in)
https://www.econtalk.org/sabine-hossenfelder-on-physics-real...
If "person who's content is available on YouTube" qualifies one as a YouTuber, then yes
But even if we want to be very general and just parse it using default English, the -er suffix would mean someone who is doing something. I believe it would be more accurate to say Feynman was YouTubed.
Even posthumously he’s been one of the most popular educators on youtube.
It seems like this is just the “you need to study the problem and it’s dependencies for many years to understand the physics and mathematical details” problem. Which does not actually seem like much of a problem to be solved, just a reality.
Your example in the other comment depends on linear algebra, so ~first year undergraduate of most technical fields. IMO that’s fundamentally more accessible than the black hole information paradox, with its dependencies on general relativity and quantum field theory.
I’m sure more technical videos and posts could be made about this, but how small would the target audience be?
Which is exactly the "draw the rest of the owl" problem. Something that takes substantial time, skill and knowledge to achieve is demonstrated as "three easy steps."
The key word is ‘numerical’.
Maybe you’ve yet to see or do numerical analysis of both general relativity and quantum field theory but they are in fact both linear algebra.
Which as you point out is “fundamentally more accessible”.
Source: I'm a Physics PhD
Now, GR or at least the core of it is a lot easier as long as you’re mathematically prepared for differential geometry.
Btw, I took particle physics from the Peskin in Peskin & Schroeder when I was an undergrad. Everyone in the class (like four of us by the third week?) was super duper lost by midterm.
The same is true for quantum field theory. I'm curious why you're so confident that they are "in fact both linear algebra"
When talking to a musical educator complaining about struggling students just starting out I’ll say “I found it extremely helpful to paint the C major scale on my guitar when learning music theory.”
And they respond “Well what about chromatic 12 pitch atonal music?! What would Anton Webern say?!”
Some subjects are just plain hard and only accessible to smart persistent people.
GR and QFT definitely qualify - but they seem to be warm up exercises compared to whatever Quantum Gravity will eventually become.
I call you out. Don’t think you’ve done either one after reading your comments.
His yt channel is an excellent resource of theoretical understanding through numerical methods.
Here’s my favorite series where he confirms/elucidates a “strange” theoretical fact with an explorative numerical style: https://m.youtube.com/watch?v=byjhpzEoXFs
Simply superb content.
While I also appreciate a healthy dose of skepticism, and in weaker moments I can also get very cynical, this is not a very constructive attitude to science or life in general. If we were all doing science like that, we'd not have gotten anywhere.
As mentioned elsewhere in this discussion, a "solution" will not only consist of a theory that "solves" the problem in the sense that Sabine described, but even propose experiments that can actually be executed within a time frame (and budget) in order to validate the theory. Just because her colleagues (and Sabine herself) have failed at that so far should not be a reason to give up. With that attitude, there wouldn't be much science left today.
What I read in this article is pointing out that these black hole information loss papers are not doing science.
There are no falsifiable predictions, because we can't actually run a meaningful experiment on black holes. At least in regard to information loss. As someone else in the thread said, thinking about how things might work without a way to tell if your idea is right is philisophy. Last I heard, physics was supposed to be science and not philisophy.
"Not falsifiable" commonly refers to no experiment being possible in principle. That's quite different from what you mean.
It seems like her impression is that (parts of) the high energy theory community are too optimistic to an extent where they unintentionally or intentionally deceive the public (who are ultimately funding science through taxes).
If that's correct, perhaps it can be beneficial for science as a whole to provide a counterbalance?
This is a fair take. I really appreciate it. Sometimes i'm too cynical about the motivations of the folks being cynical :)
This has certainly been my experience in cryptography: as a student you often start off proposing schemes which get broken by your advisor/collaborators (not out of malice, but just because broken schemes are broken), and so you learn to react to novel ideas with "how is this broken?". However, as you mature, you realize that all new correct ideas arise from the ashes of many broken attempts, and so your reaction slowly changes to "how can I fix this?", leading to an overall more positive outlook on both your work as well as others' work.
But that’s not what most scientists do today. They are cynics masquerading (even to themselves) as optimists. They have preprogrammed themselves to never even think of a question that has a good chance of failing, modern academia has collectively programmed them all to only ask questions that never have a real chance of being false to begin with. So just softball niche questions or in the case of the videos topic, reformulate the question in a way so that the answer doesn’t fundamentally solve the real problem. Both because the real problem might be unsolvable and also because if you solve the problem then you have fired yourself from yoUr job.
Now you might think I and Sabine and others are just shitting on scientists doing the work, but many of us are only doing so after wasting decades with this establishment and giving up. Perhaps you can see that for yourself earlier and save yourself a lifetime.
I don't think Sabine is telling us to "give up". Let me give an example. Suppose there is an experiment that is solvable by building a super collider whose diameter is the diameter of the moon. Should we be rushing to fund it now? Or should we wait for lunar construction techniques to bring the costs down first?
Even a lot of physics myths around doing the hard thing to seed other tech is flat out wrong. There's an often claim that the ssc tech resulted in MRI superconducting magnets; but that tech was actually descended from NMR research done by petrochem (and if you can believe it, dairy) industries that precedes the SSC by 1.5 to 2 decades.
I think Sabine has settled into a niche with a few other physicists of "If you don't even have a hope of testing your theory, what's the point? Let's direct those resources into things that can be tested sometime in the next century."
Theoretical physics seems to have a few arenas where it has just completely left the realm of reality. It's not even like "Well, maybe we can test these in a few years when technology gets better" but more like "We would need 15 orders of magnitude more energy than exists in the universe to test this."
Physicists for the most part aren't talking down spiritually, they're just talking to the equivalent of a child.
I think there's a General Relativity book that does the same thing only with the schwarzchild solution, but I'm not sure who wrote it.
I think it’s a paradox to say “you need to be fully educated on a subject to engage with an explanation intended for lay people”.
Some of these channels have millions of viewers, and the one I think you are referring to has 11M. Remember when the Reimann Zeta function summation of -1/12 was all the rage, or the Banach-Tarsky paradox? Google either of those and there are a dozen YouTube videos for each one. These are post-doctoral topics that become name-dropping buzzwords, I would never claim to understand these, yet people who watched the video do.
Which seems like a lot what the internet has given us: information, not knowledge. To quote a speaker I saw at Siggraph in ~1996, "Information is not power, knowledge is power." This hit home. We've seen it from people reading WebMD in the 2000's and then trying to teach their doctors, to the crazy scientific ignorance about mask mandates today. All because of internet edutainment or in the worst-case misinformation.
Anyway, yes, her article got cynical fast, but I think it was because she bounded with realism because plotted on the arch of technology evolution: it would take thousands of generations (even at an exponential pace) to create tech that can take actual measurements. That seems like good science to me.
* Mathologer: "Ramanujan: Making sense of 1+2+3+... = -1/12 and Co." https://www.youtube.com/watch?v=jcKRGpMiVTw
* Vsauce: "The Banach–Tarski Paradox" https://www.youtube.com/watch?v=s86-Z-CbaHA
Each one is like half an hour long, but you may need to pause them and rewatch them a few times to understand the details.
The fact that he didn't, to me, indicates he is obviously not in the business of making money.
Given measurements of the Local Standard of Rest and star surveys of the LSR neighbourhood, it is unlikely that there are any stellar-mass black holes at all within about 300 light-years of Earth (and X-Ray astronomy and microlensing/MACHO-hunting surveys each gives even more pessimistic nearby black hole numbers) so the prospects of getting a signal from an Earth probe dispatched to close to a black hole within a millennium is, pardon the pun, faint. There could be only some thousands up to a few (~10) million in the Milky Way, a galaxy of hundreds of billions of stars at this point in its evolution. There may be as many as ten billion stellar black holes in and around Milky-Andromeda five or so billion years in our future.
I note you also say "creating ones experimentally", but that is in the realm of science fiction at this stage, barely hard s.f. even for microscopic masses, and it gets decidedly soft as one climbs up the mass scale from nanograms.
Also of course experimental black hole creation is in the realm of science fiction but so was going to the moon 150 years back. Question is if there’s a fundamental theoretical block , similar to FTL , in achieving it?
Some researchers have proposed semi-seriously that there could be a small black hole lurking at the outer margins of our solar system. That obviously could be reached within a thousand years in principle (depends on how far out it is, although it must be considerably beyond Neptune's orbit), but we are perhaps decades out from a detection of it, if it even exists. The idea is, however, one of many reasons to build telescopes that are mainly to be aimed at objects in our star's immediate neighbourhood, to do very sensitive mapping of the metric the solar system sources (or, in Newtonian terms, our sun's gravitational field and the disturbances planets, asteroids, and other objects impose on it), and to do "nanolensing" observations on objects at a variety of distances. The Vera C. Rubin Observatory's Legacy Survey of Space and Time is a step in that direction, and will eliminate many possible hiding spaces for a nearby black hole.
There is an upper limit on proper acceleration proportional to mc^3/h-bar for massive subatomic particles like electrons, from Caianiello (1984) <https://doi.org/10.1007/BF02748378> (free PDF <https://link.springer.com/content/pdf/10.1007/BF02748378.pdf>) arising from uncertainty relations. We have no hope of producing that kind of acceleration as even with the best possible superconductors an Earth-bound accelerator can manage much less than a millionth of that <https://arxiv.org/abs/quant-ph/0407115>, and we need to be much closer in order to make a black hole out of small charged particles. Large particles are even harder to accelerate extremely. We also have a power-density problem: what fuels such an accelerator? Alternatively, accelerating less extremely means much longer linear accelerators, or nearly equivalently, less curvature in a ring accelerator. I think we're looking at a ring about the diameter of the orbit of Jupiter, or a linear accelerator crossing a large chunk of the inner solar system, but that's just a guess (someone has probably worked out some comparable size for ultrarelativistic heavy ions (lead or gold) though).
So the tl;dr is that it's not impossible, and it's easier than trying to throw large stars at one another, but neither is really plausible within your 1000 years. And neither is easier than sending a probe to V616 Mon. and waiting several thousand years. And we might do better by making really good observatories in and around Earth. Technically we could do all three, I guess, but we're more likely to live long enough to see improvements on the results from the <https://eventhorizontelescope.org>.
And of course, what I said about your three-digit timeframe goes double (if you pardon the expression) for your two-digit one.
FWIW if I live long enough for someone to confirm that Planet X is a black hole I’ll strip naked and dance in Times Square because golly is that such an out of the blue hypothesis.
The highest energies of UHEC protons are probably not caused by "instant" accelerations as in a supernova (which certainly produces high energy protons that fly across the cosmos). Instead, they're more likely drawn up magnetically around the supermassive black holes powering active galactic nuclei (quasars, blazars and so on) and into jets, where they spend possibly weeks or accelerating through a natural counterpart to a particle accelerator which benefits from things like inverse Compton scattering (a hard X-ray from the black hole hits our proton boosting the proton's momentum and lengthening the X-ray's wavelength (i.e., reducing its momentum)).
But because we don't know OMG!'s origin, we can't say if OMG! tests any acceleration limit originating in uncertainty relations. Also, the Caianiello limit is unhealthy for protons (which OMG! likely was) staying protons: that they are composite structures makes them liable to be blown apart by extreme acceleration. Assuming it's a proton and had a blazar origin, OMG! is (really rough back-of-envelope work) likely to have fallen between 7 and 40 orders of magnitude short of the maximum (proper) acceleration. The intuition here is that OMG! is well short of Planck energy (by some 7 orders of magnitude) and super-high acceleration should take a proton from thermal energies to Planck energies. Of course a proton much closer to Planck energy could lose proportionally more along the way than a less-energetic proton, so who knows? There could be a doctoral dissertation waiting to be written on this!
Also, sorta disappointingly, OMG! could have had an electric charge greater than +1 (i.e., some ion with more than one proton) and in that case could have originated as close as our own sun <https://en.wikipedia.org/wiki/Solar_energetic_particles> <https://en.wikipedia.org/wiki/HZE_ions>.
In this paper he considers the possibility that the universe is almost dominated by tiny (~ micrograms) black holes formed in the very early universe, persisting to the 1970s and leaving tracks in sufficiently large and sufficiently numerous cloud chamber detectors given sufficient time. (This was also decades before optical detectors like at SNOLAB, but incongruously after the Nobel prize for bubble chambers, although it was on the cusp of the activation of <https://en.wikipedia.org/wiki/Gargamelle> and the Big European Bubble Chamber -- he would have been aware of how close they were to activation, and results from smaller predecessors). Fascinatingly, several years after publication, conceptually similar papers were put forward for hunts for dark matter, even though it is impossible to extract from this Hawking 1971 paper that he was thinking of galactic rotation curves.
The paper also pre-dates evidence for the acceleration of the expansion of the universe, and contains the curious sentence: "An upper bound on the number of these objects can be set from the measurments by Sandage (7) of the deceleration of the expansion of the Universe." However, the excellent Allan Sandage paper of 1961 <https://articles.adsabs.harvard.edu/cgi-bin/nph-iarticle_que...> does not really let one take sides ("It suffices to say that no uncontested evidence is presently available [to choose between steady-state and expanding universe]") like Hawking 1971 seems to. Nevertheless, Hawking raises an interesting and slightly spooky premonition of dark matter at super-galactic scale, keeping clusters of galaxies looking like clusters over long periods of time, rather than separating into individual isolated galaxies: "This extra density could stabilize clusters of galaxies which, otherwise, appear mostly not to be gravitationally bound." In context Hawking was clearly thinking about a universe that grows into some very long term non-expanding/non-contracting configuration which includes clumps of galaxies.
Hawking 1971 tl;dr is "good luck actually observing these small weakly-interacting things, if they exist". Which is appropriate for 21st century particle dark matter experimentalists too. :^)
Penultimately, some patience with google scholar has found that a small handful of the many articles citing Hawking 1971 explore the idea of primordial black holes as dark matter. (but see below).
And finally, sometimes asking questions like yours (how does one build a very small black hole?) and taking them seriously leads to some very early, vague, ideas about how the universe might work in bulk, which is probably very very similar to what provoked Hawking's letter about 52 years ago.
- --
[1] https://sci-hub.se/https://doi.org/10.1038/248030a0
The relevant part of his evaporation letter: "There might, however, be much smaller black holes which were formed by fluctuations in the early Universe². Any such black hole of mass less than 10^15 g would have evaporated by now." In that context, I think a very low motivation for someone not terribly interested in the history of primordial black hole research to have previously hunted down a 1971 note whose idea was shortly thrown under the bus by its author is fairly understandable. :^)
Hawking 1971 is however an early note on primordial black holes (the only obvious earlier one in the literature is from Zel'dovich and Novikov in 1966), and that seems to be behind many of the citations I briefly sampled. Practically all also cite the more extensive follow-up Carr & Hawking 1974 "Black holes in the early Universe" <https://ui.adsabs.harvard.edu/abs/1974MNRAS.168..399C/abstra...> which is very close in submission date to the publication date of Hawking's "Black hole explosions?" yet does not appear to consider evaporation.
There are hundreds of billions of regular stars in our galaxy (and in similar galaxies), some much more massive than neutron stars, and all of them have a much larger cross-sectional area than neutron stars. (NSes are tiny -- compare with New York City -- <https://commons.wikimedia.org/wiki/File:Neutron_Star_Manhatt...>) So small black holes are much more likely to collide with ordinary stars, and even Long Island NY and other comparable features on Earth, other solar system bodies, and exoplanets (we've counted five thousand of those now, and a lot of them are considerably closer than the nearest neutron star). But we see no evidence of that.
If they're moving gently along with all the stars in their neighbourhood, collisions between these asteroid-mass black holes and anything appreciable is unlikely (just as star-star collisions are very rare) that we wouldn't expect to see that ever happen.
Wandering pretty violently in comparison to nearby bright objects, an asteroid-sized black hole is likely to zip by a star and its satellites on a (strongly) hyperbolic orbit, like <https://en.wikipedia.org/wiki/%CA%BBOumuamua>, just dimmer. It's even less likely to hit a neutron star because of the NS's compactness, and because the geodesics it generates produces greater hyperbolicity near its surface (because that's quite near the centre of mass) than near the surface of a regular star. And hitting a sun-like star is already hard for an asteroid: <http://curious.astro.cornell.edu/our-solar-system/comets-met...> (likewise, we have never seen a small black hole flare up near our sun).
It is more likely that -- very occasionally -- an asteroid mass object (by strong equivalence principle, it does not matter if it is rocky, icy, or a black hole) is captured into a weakly hyperbolic orbit, or even an elliptical one, by a ~stellar mass. There are reasons to think an NS might be more efficient at this than a fatter star. But then you get a (possibly wide) binary star-BH that soon becomes hard to distinguish from a binary that formed together and "grew up" as stars together, with one happening to get too fat and collapse. The missing nebula in the captured-BH case would raise questions.
I don't buy the idea of a small black hole that somehow manages to slice through a neutron star doing much to the neutron star. Typically many many solar masses of matter fall back onto neutron stars after they violently form, and almost certainly there are asteroid-sized chunks of heavy atomic nuclei landing on them often. If you had some cosmic pool shark aim a small BH so precisely that it not only enters the (very very very very very small) NS core, but stays there, coming to a stop just so, the NS would collapse into the BH [1], because no matter can rest just outside the event horizon. The inner hard "platform" matter would fall in first, and in turn the freshly unsupported upper layers which had been "standing" on that platform would fall in, and so on down the widening drain. But if you shoot an asteroid-mass through an NS, it's hard to think there'd be much consequence. Sufficently slowly, there could be a nuclear explosion but probably not one that's catastrophic to the NS's integrity: I think it'd be hard to distinguish from the nuclear explosions from the rain of supernova-remnant material falling back down onto it, or random interstellar matter an isolated NS might encounter, or crustal quakes.
(I could be convinced by an academic paper equipped with an NS equation of state though (see [1] again), in principle, but very likely not a youtube video, much less a second-hand summary of one :-) ).
To a zeroth order approximation, the effect on a planet of a collision with an asteroid-mass black hole is not going to too different from the effect of a pair of collisions with ordinary asteroids (the "pair" is from the ingress and egress sides of the "tunnel": actual asteroids will tend hit a surface and stay there, or break up in a thick planetary atmosphere if there is one, etc., while small black holes will sail through). The asteroid-mass black hole will certainly create lots of electromagnetic interaction on contact; you can think of how black holes tend to have bright accretion structures, so you expect a big flashy kaboom as with a rock (but with more X-rays and gammas).
While we can resolve exoplanets, we can't quite see them being hit by asteroids, assuming the Jupiter-type exoplanets are hit with some frequency, we could blame limits in our telescopes. (JWST will help very soon!). Rockier exoplanets are no more likely to be hit by asteroids and lit up brightly than Earth or Mars. Asteroids are just tiny masses compared to these bodies.
- --
[1] fate of a neutron star with an "endoparasitic" small (and possibly primordial) black hole, that "just so" is already inside the NS: https://arxiv.org/abs/2101.12220
Of course you can't quickly teach the general public the actual math of quantum mechanics or whatever, but I think it's far far better to explain things in simple terms without further complicating the subject with awkward analogies.
You cannot teach them that in any acceptable length of time. Just showing code and pointing at various parts: "this does x", "this other thing does y" is no more useful than using abstract art: the subtle syntax and semantics you are so accustomed to interpreting that you do not even notice it (array subscripts, point dereferences, etc.), they cannot even realize is there.
This is why science educators don't show the math. They're not talking down because they want to, but for the same reason you can't explain object relational impedance mismatch to a 7 year old.
> Classic Poisoning the Well fallacy. The audience will be predisposed to compare the hypothesis to religion and not take it seriously before any argument is even made.
Why shouldn't you take religion seriously? Pretty rude to religions.
> There are many theories in theoretical physics where it is very difficult or impossible to come up with experiments to prove or disprove them, e.g. Everett’s Many-Worlds Interpretation or the String Theory. One might call these sorts of theories metaphysical, fine, but it doesn’t make them into disproven pseudoscience or religions.
On the other hand, yes it does do that, which is one of her most common points.
> Again, it doesn’t mean much that nobody currently knows how to do it. No practical quantum computers exist yet, so obviously, nobody tried to run anything useful on them and more quantum algorithms will be found that run with lower time complexities than on classical computers.
This doesn't help him because there aren't any algorithms that are uncomputable on classical computers but computable on quantum computers. She's also correct than quantum computers are special-purpose things and otherwise fairly useless.
Though I don't know why the universe wouldn't be a computable function if you allow the computer to have a true random number generator.
More importantly, I know that believing in superintelligence means you’ve been reading too much SF and have forgotten that things can’t exist just because you can imagine them existing. Spend any time worrying about that and next thing you know you’ll have moved to a group home in Berkeley and joined an effective altruism cult.
Why would human intelligence be the absolute limit of what is possible?
Seems implausible.
Or maybe you mean something else by superintelligence, like, "more intelligent than humans in a self-improving way, and then FOOM and then it is necessarily more powerful than everything else", in which case, yeah that might be impossible (the upper bound of intelligence could perhaps be, while still higher than human intelligence, not actually all that much larger)
I'm willing to accept that intelligence exists insofar as it's the difference between a human and a gorilla. But is that something you can have "more" of?
That concept implies humans are smarter just because their brains have more IQ, but maybe it's a single fixed function - so the gorilla has 0 intelligences and the human has 1 - then something with 2 isn't a superintelligence, it's more like a conjoined twin and would just find it wants to do two things at once and can't.
But maybe superintelligence means someone who can think more quickly than you. So maybe that's about as scary as Magnus Carlsen or Terry Tao, but outside chess we already have calculators, and people who do math by hand don't seem to fear calculator users as superintelligences despite their superior results. In the real physical world, thinking faster doesn't necessarily get you better results because you still have to be correct, which means guessing and checking, so maybe it's the super-patient person we should be afraid of.
Admittedly neither of those are proofs it's an impossible concept, just some plausible alternatives.
> Or maybe you mean something else by superintelligence, like, "more intelligent than humans in a self-improving way, and then FOOM and then it is necessarily more powerful than everything else"
Yes, I believe this scenario they're worried about is impossible because something will always appear that prevents it, and they've left all possible somethings out of their scenarios because they are only imagining them and not actually testing them.
Such things would be entropy, energy consumption, communications delays, self-interest, "self-improving" turning out impossible due to Goodhart's law, etc. Examples of it not happening are large corporations (which always eventually stop growing, need constantly increasing inputs, and are actually composed of lots of smaller intelligences that don't all have the same goals), and the Mind AIs in Culture books (which have to be held back to even want to think about reality, and if they get too smart immediately stop caring about reality and just think about math forever).
Thinking faster isn't synonymous with being more likely to take risky mental shortcuts.
Imagine a person who, for every hour for us, experiences an entire year's worth of time (and they can make/review notes on their ideas as quickly, compared to their experience of time, as we can compared to our experience of time. So they aren't limited to being able to keep a small amount of information in working memory.).
Would we not expect such a person to be able to consistently outwit us?
It seems this is enough to demonstrate it is at least a coherent concept, even if something as extreme as what I just described might not be physically possible.
(Of course, talking about subjective experience in this way might be a little bit of a distraction, as perhaps there is no guarantee that a highly intelligent (in the sense of "able to formulate plans and such in the world, etc. ) agent have any internal experience. But, it is at least easy to imagine. And, hey, maybe it does happen to be the case that any highly intelligent agent would have internal experience, idk.)
But, even if something that extreme (able to do as much planning and reasoning over the course of an hour as one of us could do in a year) is physically impossible (and I wouldn't be too surprised if it is physically impossible), I would still expect that there are things at least a little bit in that direction from us. Which, yes, might be be basically along the lines of [ (you or I) : Terry Tao :: Terry Tao : (hypothetical agent) ] (or perhaps iterating that analogy a few times, idk. Again, not sure how far from the limit humans are, it just seems implausible that humans are at the limit.) .
And, if in the same sense that Terry Tao is more intelligent than I am, there were an agent which was more intelligent than all (individual) humans, and wasn't guaranteed to have human-like goals/values, nor to value human interests highly, then yes, I do think that this could be rather concerning, depending on the size of the difference in intelligence, combined with what the strategic positions are.
> and people who do math by hand don't seem to fear calculator users as superintelligences despite their superior results.
This doesn't really seem like a serious argument to me?
Like, obviously people who don't use calculators, can, just, procure a calculator if they find lacking one has put them at a disadvantage? It seems a silly argument.
Regarding FOOM and such, yes, I'm not particularly worried about it, or in fact about AGI at all?
Also, the example of the Mind AIs in the culture books, is not evidence, because that is a work of fiction? The world is not to obligated to follow tropes from stories we tell, and stories tend to differ from reality in ways that make them better stories.
Have you tried arguing both sides of the point you are arguing?
It took me a while to realize why I thought everything she said sounded so authoritative :)
In a world where FUD spreads faster than truth and not everyone is an expert in every topic: her work does as much, if not more, harm than good. Particularly in fields she is not an expert in.
I like Sabine’s personality and the fact she stands in defiance of the status quo of scientific outreach, namely my two least favorite tropes: “science is fun” and “let me explain something using terrible metaphors because I fail to understand the math myself”.
Sabine, if you’re reading this… please create more technical content.
Maybe a companion video to your more general takes.
One that shows the numerical side of things. Your experience and personality already lend itself to this effort.
You say “the math is insufficient” show us the math! Show us how it’s insufficient using numerical examples.
Scientific outreach has a real “draw the rest of the owl” problem and I think Sabine is perfectly poised to fill the gap.
There’s a George Carlin quote I’ll paraphrase ~“never talk down to your audience, they’ll catch up eventually.”