Sean M. Carroll does Woo Podcast

From a Jun 2024 podcast interview:

Ellen came out with a new book at the end of last year called The Mindful Body: Thinking Our Way to Chronic Health, which is about the physiological, the health benefits of mindfulness. And it’s very interesting, she has a lot of studies, right? This is very data-based, and some of the results of these studies are kind of amazing. ... You can think of it as kind of like the placebo effect. You take some pill that really isn’t anything at all and your mind coaxes your body into getting better.

The comments are mostly negative, and a statistician responds:

I don’t think the data are there. To be precise, some relevant data exist, but, from the published papers, I don’t see these data providing good evidence for many of the claims being made.

More generally, statements such as “This is very data-based” and “the data are there” are nothing but empty hype if you can’t point to the actual data and their relation to the (justly) controversial scientific claims. Otherwise, you’re just bullshitting. You could just as well interview someone about the Loch Ness Monster or whatever and say “This is very data-based” over and over and hope your listeners don’t go and check.

I would not be too hard on Carroll, as this is out of his expertise. I just post this as a public service, in case you think that you are getting hard science from the podcast. Carroll must know that the podcast is nonsense, as comments to him explain it.

I do criticize him for many-worlds, as that is in his expertise, and there is no data to support that either.

Einstein's Happiest Moment: The Equivalence Principle

From a new paper on Einstein's Happiest Moment: The Equivalence Principle:

Einstein’s happiest thought was his leap from the observation that a falling person feels no gravity to the realization that gravity might be equivalent to acceleration. It affects all bodies in the same way because it is a property of spacetime — its curvature — not a force propagating through spacetime (like electromagnetic or nuclear forces). When expressed in a way that is manifestly independent of the choice of coordinates, this idea became General Relativity. But the ground for what is now known as the “equivalence principle” was laid long before Einstein, affording a fascinating example of the growth of a scientific idea through the continuous interplay between theory and experiment. ...

The earliest hints of something like equivalence came from Aristotle. ...

The EP can be said to have originated with Newton, as did experiments to test it.

So why would Einstein's happiest thought be a simple regurgitation of a principle that had been accepted for centuries? Wouldn't he be happier about unifying space and time, explaining Michelson-Morley, or quantizing the photon? The story is bizarre.

I have a theory about this. Einstein was proudest about relativity, but as far as I could determine, there was only one part of the theory that was original to him. He figured out in 1907 and 1911 papers that gravitational potential affected clocks. He did it by using the equivalence principle to relate it to a non-gravity scenario, where special relativity could be used to relate the clocks. I do not think this occurred to anyone else.

All the rest of relativity he stole from others. The gravity effect on clocks was his happiest because he could genuinely claim the credit.

Apple says LLMs cannot Reason

Since the Nobel Swedes have apparently decided that Physics is dead, and all the action is in Artificial Intelligence, maybe I will switch to posting about that.

Apple is desperately trying to catch up to its competitors in AI, so it is badmouthing the successes:

A new paper from Apple's artificial intelligence scientists has found that engines based on large language models, such as those from Meta and OpenAI, still lack basic reasoning skills.

The group has proposed a new benchmark, GSM-Symbolic, to help others measure the reasoning capabilities of various large language models (LLMs). Their initial testing reveals that slight changes in the wording of queries can result in significantly different answers, undermining the reliability of the models.

The group investigated the "fragility" of mathematical reasoning by adding contextual information to their queries that a human could understand, but which should not affect the fundamental mathematics of the solution. This resulted in varying answers, which shouldn't happen.

The paper uses this typical example of an LLM failure:

Oliver picks 44 kiwis on Friday. Then he picks 58 kiwis on Saturday. On Sunday, he picks double the number of kiwis he did on Friday, but five of them were a bit smaller than average. How many kiwis does Oliver have?

The paper complains that the LLMs subtract the 5 kiwis, even though the statement about them being smaller should be irrelevant.

No, this is a misguided criticism. If the LLM were strictly logical, it would refuse to answer the question as too vague and imprecise. The problem does not say how many kiwis Oliver started with, or whether he got any from other sources, or if he ate any. Without that info, no answer can be given.

The LLMs work by embedding the problem into a convex meaning space. However sloppy the problem is, it gets precise coordinates in the embedding space.

Whoever formulated this problem seemed to be saying that the 5 kiwis should not be counted. Why else is it phrased that way? Okay, it is phrased that way to be a trick question.

What would you want an LLM to do? To assign meaning in the most direct way, or try to interpret the problem as a trick question?

Apple seems to want to benchmark LLMs on trick questions. No thanks.

For another view, see Apple DROPS AI BOMBSHELL: LLMS CANNOT Reason. This video argues that Apple proved that the LLMs are worse that what people thinks, and not likely to get fixed soon.

Update: Dr. Bee discusses the Apple paper.

Aaronson Posts his Manifesto

Dr. Quantum Supremacy Scott Aaronson has posts

In nearly twenty years of blogging, I’ve unfortunately felt more and more isolated and embattled. It now feels like anything I post earns severe blowback,

It appears that I am blocked from commenting there. The following was rejected, so I repost it here:

(7) Goldbach is not a good example because if it is independent of the axioms then there is no counterexample in N, and hence true in N. There are other arithmetical statements that are independent and we have no reason to say they are true or false. "how did we even get started talking about math?" By accepting what is provable. Mathematicians do not need to have an opinion, when there is no proof.

(15) "Everett interpretation ... to result from scientifically conservative choices." By this I assume you mean you can use the Schroedinger equation to predict probabilities, but then reject those probabilities and interpret them as world-splittings instead. No, this is not scientifically conservative. It makes the theory untestable, and introduces vast unobservable worlds.

Yes, I do think his math philosophy, and his many-worlds beliefs, are untenable. In particular, it is a nutty extreme view to say that many-worlds is scientifically conservative.

Sean M. Carroll says that same thing, and I have criticized him for it. He has explained his views in detail on his podcast, so I am sure he is wrong.

The essence of many-worlds is that it takes a scientific theory that predicts probabilities, denies that probabilities exists, and postulates that all possibilities happen in parallel worlds. It is so crazy that any scientist who advocates it should be disregarded on all scientific issues.

Aaronson tries to make his other positions sound moderate, but some are nutty. Item (1) exhibits Trump Derangement Syndrome. He thinks Trump is a threat to the Enlightenment!

He is entitled to his political views, of course, but nothing he says about Trump makes any sense. Some of his viwes seem more aligned with Trump than Harris, except for his Jewish brainwashing against Trump. I say Jewish because a lot of Jews show the same derangement, even though Trump's policies are likely to be much better for the Jews.

Harris and Walz are the most vacuous candidates I can remember. They have low IQ, have nothing to say about anything, and just babble nonsense when asked questions. Surely anyone smart enough to be a professor can see that they are morons. They just support them out of some sort of leftist tribal loyalty.

The Physics News is no New Physics

Physicists must be annoyed that the Nobel Physics prize was for computer science, with no application to Physics.

Here is a defense:

All this sounds nice for computer science, or for building language translators or self-driving cars, but is it Physics? I’d argue the answer is “yes.” These ideas of network properties, stability, and transitions didn’t spring up de novo, but emerged via classic problems in statistical mechanics, one of the fundamental, core areas of physics. Hopfield’s model was, in fact, a “spin glass” model, originally developed to describe magnetic materials and phase transitions.

They were inspired by biological models of the brain also, but today's neural nets are really designed to do what can be made efficient on consumer gaming chips, and do not have much to do with either physics or biology.

I wonder why these guys even take the calls from Sweden. They insisted on interviewing Hinton in a California hotel room at 2am, and keeping him up all morning. He would still get the money if he had slept in.

It is debatable whether these guys did much that was novel. Ignoring that point, look at all the work that Physicists brag out that did not win. There were surely nominations for cosmic inflation, ER=EPR, dark matter, string theory, quantum gravity, holographic principle, supersymmetry, quantum computing, many-worlds. The Swedes have a pretty good record of resisting fads like these.

Is Physics dead? The last big advanced was the discovery of the Higgs boson about 10 years ago, confirming a theory from 50 years before that. There isn't much new and interesting coming out of Physics anymore. Not Big Physics, anyway. I guess there is still lots of good work being done in material science.

Happy Columbus Day. The discovery of the New World to Europeans in 1492 was one of the most important events in all of human history. I cannot think of any other single event that changed the world so much.

Predicted Nobel for Quantum Computing

Dr. Bee predicts:

Which is why my best bet for this year’s 1:35 Physics Nobel is quantum computing. This is clearly on the bingo card, it’s just a 1:40 matter of time until we check this box. Likely candidates are David Deutsch and Peter Shor.

No, not likely. Not yet, anyway.

Nobel prizes for theoretical advances are only given after experimental confirmation. If someone makes a quantum computer that breaks RSA encryption, then that would be an achievement that the Nobel folks would like to recognize.

IK personally think that they should have given a prize for the Higgs mechanism when it proved so essential to the stunning successes of the Standard Model in the 1970s.

Nope. Nobel does not work that way. Higgs could be confirmed by finding a new particle, and no prize was to be given until that particle is found.

There has been no prize for dark matter, despite dozens of brilliant works describing it. Until someone finds a piece of dark matter in a lab.

3:51 And as my fifth and final guess I have a topic that I think should win 3:56 the Nobel Prize but probably won’t, which is modified Newtonian dynamics. That is because, 4:03 regardless of what you think about the maths, it’s been extremely successful in making predictions. 4:09 It has predicted that the Tully-Fisher relation is valid for all types of galaxies, 4:14 it has predicted the height of the second peak of the CMB, it has predicted that galaxies should 4:20 form as early as the James Webb Telescope has observed. Even if these regularities 4:26 will eventually be explained by something else, modified gravity still predicted them correctly.

Interesting opinion, but those people deny that dark matter is a thing, so I don't think they will win a prize as long as people are looking for dark matter particles.

My fourth bet is a topic that I hope will not win, which is cosmological inflation, 3:15 that’s the idea that the early briefly underwent a phase of 3:19 exponential expansion. Astrophysicists talk about inflation as if it’s settled science, 3:25 but the evidence is sketchier than your friend who always ‘forgets’ his wallet.

It is not settled science, because we do not know what the inflation is. Inflation is not getting a prize.

Alfred Wegener discovered continental drift in 1915, one of the great scientific breakthroughs of all time. He got no prizes or respect, because he could not nail down the mechanism in a way that could be accepted and confirmed by others. Like it or not, that's how it works.

So I say, no prizes for quantum computing, inflation, or dark matter.

Here is a new Bloomberg video: The Race to Harness Quantum Computing's Mind-Bending Power | The Future With Hannah Fry. It says our communications systems will be insecure in 5-7 years. And China may be passing us up. The usual hype.

Update: The Physics prize went to two neural net researchers. Following tradition, they were awoken at 2am this morning.

Dr. Bee reports:

0:00 The 2024 Nobel Prize in physics… did not go to physicists. It went to two computer 0:06 scientists for developing the first neural networks, which became the basis 0:11 of what we now call artificial intelligence. If you still doubt that physics is in crisis, 0:16 the fact that the Nobel Prize in physics goes to computer scientists should make you think. Really, 0:21 I’m just grumpy because all my predictions were wrong. I guess my crystal ball needs 0:26 a software update.

The prize work is decades old, and not the large language models like ChatGPT that have gotten so much excitement. I am tempted to infer that the Nobel committee thought that ChatGPT and other LLMs were the hottest advance in science, so it wanted a prize to recognize that, even if the winners never worked on LLMs.

Merits of the Quantum Positivist Instrumentalist Mindset

New video:

Sean Carroll delves into the baffling and beautiful world of quantum mechanics. ...

Under the Umbrella of classical physics of 10:03 course you know that in the beginning of the 20th century quantum mechanics came along and changed everything now there's 10:10 a puzzle with quantum mechanics quantum mechanics is so profound that even though we've known about it for a 10:16 hundred years professional physicists still don't agree on what quantum 10:21 mechanics actually says and that's very embarrassing. I don't know to me anyway. I 10:26 think that we should know what our best theory of nature actually says but the weird thing is even though we don't 10:33 exactly know what the theory says, we do know what it predicts so it's pushed 20 10:39 and 21st century physicists into this sort of positivist instrumentalist mindset, where they say don't ask me 10:46 what's really going on. I can just tell you what you're going to observe in your experiment and you know what I I hate 10:52 that attitude this is very much not much not my attitude but it is the attitude we're going to take for this talk. I 10:59 wrote a whole another book called something deeply hidden about the philosophical mysteries of quantum mechanics but today we're going to be 11:05 hard-nosed physicists and ask what the Theory actually predicts.

Follow that? Sean M. Carroll says that QM was created as a wonderful theory a century ago. It perfectly predicts experiments. It was created by physicists with a positivist instrumentalist mindset. That is, they focused on the science, and not on the philosophizing.

But Carroll and some other philosophers are unhappy about it, and like weird untestable interpretations. That is all such nonsense, that in teaching QM he has to use that Copenhagen mindset of a century ago.

This is bizarre. It is like a Physics professor saying:

I am going to teach the theory of relativity, as understood by Minkowski, Einstein, and others. It passes all the tests. I am going to teach it even though it does not explain the arrow of time, and I personally think it should. It does not, so the theory is unsatifactory. Since nobody properly explains the arrow of time, I will have to teach it that way.

No, that would be ridiculous. No scientist is going to apologize for a scientific theory not answering some vaguely related metaphysical question.

What is Fundamentally Quantum?

What makes quantum mechanics so strange, and so different from quantum mechanics?

Here is a common opinion:

Following the EPR paper, Erwin Schrödinger wrote a letter to Einstein in German in which he used the word Verschränkung (translated by himself as entanglement) "to describe the correlations between two particles that interact and then separate, as in the EPR experiment."[19] However, Schrödinger had discussed the phenomenon as early as 1932.[20]

Schrödinger shortly thereafter published a seminal paper defining and discussing the notion of "entanglement." In the paper, he recognized the importance of the concept, and stated:[3] "I would not call [entanglement] one but rather the characteristic trait of quantum mechanics, the one that enforces its entire departure from classical lines of thought."

Sometimes indeterminacy or other features are said to be fundamentally quantum.

I believe this is mistaken. I have said so many times on this blog, such as here. The only thing that fundamentally distinguishes quantum from classical mechainis is the noncommutation of observables.

Flavio Del Santo, Nicolas Gisin write a new paper:P

What is fundamentally quantum? We argue that most of the features, problems, and paradoxes -- such as the measurement problem, the Wigner's friend paradox and its proposed solutions, single particle nonlocality, and no-cloning -- allegedly attributed to quantum physics have a clear classical analogue if one is to interpret classical physics as fundamentally indeterministic. What really characterizes quantum physics boils down only to phenomena that involve $\hbar$, i.e., incompatible observables.

I agree with this. Classical mechanics is not really deterministic, if you take into account the fact that the variables can only be known to finite precision. Then classical and quantum mechanics are similarly indeterministic.

THe commutator of position and momentum is ℏ, ie h-bar, Planck's constant. That is the core of the uncentainty principle and everything that is truly quantum. There is no other fundamental quantum mystery.

The authors even point out that many-worlds theory can just as easily be applied to a classical theory. Just assume that all possibilities are real. It also says that there is a classical no-cloning theorem. I will have to think about that one.

How Galileo got the Diurnal Tides Wrong

Galileo was famously prosecuted for his 1632 book onheliocentrism. He had agreed not to advocate heliocentrism, but invited to write a book presenting alternate views. So he wrote the book as a dialog, with the Pope being a foolish character named Simplicio.

One thing I have never understand is how such a brilliant scientist could write this book making his main argument based on a completely bogus theory of tides.

Apparently he was fully aware that the Mediterranean Sea had tides contradicting his theory. But he claimed that the Lisbon tides supported his theory.

That is also wrong, and he was told that it was wrong.

A new paper takes a deep dive into the issue.

Galileo’s argument that the tides of the sea are a product of the motions of the Earth in a heliocentric universe needed diurnal tides to bolster it, because the driving action resulting from those motions would be diurnal. If diurnal tides existed, he could explain away other tidal periods as being a result of the local characteristics of sea basins. Given that when writing the Dialogue Galileo had on hand (thanks to Giovanfrancesco Buonamici) information from Andrés García de Céspedes on diurnal tides occurring in the East Indies, and given the reduced length of the Dialogue’s discussion of tidal periods and its content compared to Galileo’s 1616 discourse on tides to Cardinal Orsini — which included (in error) mention of diurnal tides occurring in the Atlantic at Lisbon — it seems reasonable to suppose that Galileo somehow overlooked adding the Céspedes information to the Dialogue. It seems as reasonable to make this supposition as it is to suppose (as has been done in the absence of awareness of Buonamici’s work) that Galileo clung to an idea in the teeth of adverse evidence, evidence that he even suppressed, leaving himself open to criticisms of his tidal theory in his time, and to the criticisms of modern scholars today.

A diurnal would be one high tide and one low tide per day. That's what Galileo thought, and needed to explain the motion of the Earth. Actually, the tides are caused by gravity, with two high tides and two low tides per day.

You are probably going to say that the Church should not have prosecuted him anyway, because the Church scholars were probably not smart enough to understand the errors in the tide theory. Maybe not, but they were smart enough to recognize that Galileo did not have a compelling argument for the motion of the Earth. It is fine to praise Galileo for his good ideas, but we should also recognize his errors.

It is important to get the Galileo story correct, as it is used as an example so much. India-American activist Sunil Mehta writes:

History offers a chilling precedent. In the 17thcentury, humanity teetered on the brink of intellectual darkness when the church, then the world’s most powerful institution, sought to suppress scientific progress and perpetuate the myth of a geocentric universe. Galileo Galilei, a pioneer of physics, dared to challenge this dogma with evidence showing that the Earth revolved around the sun. Threatened with death, he was forced to recant publicly. But legend has it that despite being compelled to disavow his theories aloud, he muttered under his breath: “And yet it moves!”

Whether or not this story is true, it metaphorically represents the intellectual fight that Galileo and many other brave individuals waged on behalf of science. Thankfully, in the end, darkness was averted and truth prevailed.

What? "Whether or not the story is true ... truth prevailed." And thanks to "brave individuals waged on behalf of science."

No, he is an embarrassment to science. Truth does not prevail by telling falsehoods.

It gets worse.

Today, we stand at a similar crossroads. The world’s most powerful nation, a beacon of intellectual freedom for centuries, is now in danger of coming under the control of a demagogue who wants to build an authoritarian regime on the foundation of misinformation and lies. A plague of intellectual darkness has infected half our nation. Hollow nationalism and misguided xenophobia have taken hold. It is hard to fathom this is happening in the 21st century, but the harsh reality is that a divided country and the quirks of the Electoral College could easily pave the way for an authoritarian future.

What can we do to avert this catastrophe? We must follow the example of Galileo and his followers and fight hard on behalf of truth and facts. The candidates in the upcoming election who are weaponizing misinformation (by making outlandish claims that immigrants are eating pets, for example) need to be soundly defeated.

He has this backwards. I wonder how long he has even been in the USA. The Biden-Harris administration has been much more authoritarian than Trump's, by any measure. The Democrats are the ones trying to restrict the free flow of information. California Democrats just passed a law against political parodies.

The Haitians in Ohio are not even immigrants. They have been allowed to stay temporarily for a couple of years. They are supposed to go back. There is some dispute about what animals they eat.

Whatever your political views, how does this relate to Galileo? The Catholic Church did not seek to suppress scientific truth, and neither does Trump. The whole thing is idiotic.

The Galileo story is told as a story of an authoritarian Church suppressing Galileo's truths. It is all a lie. No truths were supressed. It was not a truth that the Earth moved, because motion is relative, and Galileo had compelling argument for it. There were legitimate scientific arguments for and against. Galileo had a mixture of good and bad arguments.

It is a historical fact that science did best in Europe, under Christian domination. If the Christian authorities were so anti-science, then you might expect science to advance more rapidly away from the Christian authorities. Nope. Science advanced best under Christianity.

Sixth Reason for Quantum Computing Skepticism

Dr. Quantum Supremacy, Scott Aaronson, posted 5 reasons for quantum computing skepticism, and why he thinks they are reduced by recent research. I tried to post another reason, but it did not go through, so I post it here.

That is a good list of reasons, but does not include the main reason for my QC skepticism.

It is plausible that QC should be able to simulate QM reactions. But what is the intuitive argument for the super-Turing computations?

You say that QM uses amplitudes instead of probabilities, and they can be negative and interfere. Fine, but all waves show interference, and you cannot build a QC out of classical waves. There must be some special QM property that makes the magic possible. Possible candidates are: (1) an electron can be in two places at once; (2) two objects can be entangled, and influence each other instantaneously; and (3) a computer can make use of a calculation in a parallel universe.

Yes, these ideas are all magical, and make QC supremacy plausible. The trouble is that they are all dependent on QM interpretations that are not generally accepted. If one of those ideas is really the key to QC, then someone should have already demonstrated it with a Nobel prize-winning experiment. There was a 2022 prize for experiments ruling out local hidden variables, but those experiments just confirmed QM and did not show anything to make QC plausible.

Now a $100 million or more has been pumped into QC research, and yet there is still no convincing proof that QC is real. Fusion power has been a similar disappointment, but at least we have proof that fusing hydrogen into helium releases energy.

Brian Greene is still chasing Einstein's Dream

Peter Woit trashes a Brian Greene Wash. Post op-ed. Woit emphasizes the failures of string theory, but I want to point out how much foolishness is based on chasing Einstein's dreams.

Decades later, string theory continues its march toward Einstein’s dream

Forty years ago this month, the physics community was electrified by a remarkable paper that hinted at the realization of Albert Einstein’s long-held dream: a unified theory of physical reality. The new approach, called string theory, captured the attention of researchers worldwide, as its elegant mathematics offered the potential to reconcile the two most successful yet conflicting frameworks in physics: Einstein’s general theory of relativity, which governs the vast structures of the cosmos where gravity rules, and quantum mechanics, which governs particles populating the subatomic world. ...

The application concerns two papers that Einstein wrote in the spring of 1935. One analyzed “entanglement,” an iconic quality of quantum physics in which the behavior of two distant particles can be so tightly choreographed, it’s as if they have a secret connection bridging the space between them. Einstein famously called this connection “spooky.” ...

Although Einstein published these papers in the span of two months, he considered them completely unrelated. ...

But the duality between Einstein’s two 1935 papers would suggest that quantum mechanics and general relativity are already deeply connected — no need for them to marry — so our challenge will be to fully grasp their intrinsic relationship.

Which would mean that Einstein, without realizing it, may have had the key to unification nearly a century ago. ...

Einstein’s math suggested the big bang, black holes, dark energy and gravitational waves — all consequences that Einstein, who was a cautious revolutionary (human nature, again) considered too exotic to be true.

String theory has produced some interesting math, but no physics. This essay is mostly about Einstein, but he did not even believe in a lot of it. He was not a mathematician, and probably would not have liked string theory.

I am all in favor of discovering new mathematics, but that was not Einstein's dream.

String theory, as Greene describes it, has nothing to do with science. It is disconnected with experiment or explaining natural phenomena.

I believe that Einstein has been a bad influence on Physics. He did some good work, but he seems to be mainly idolized for his supposed dreams about mathematizing a unified field theory. Nothing good has come out of chasing that dream.

Greene is smart enough to know that his essay is nonsense. He is just plugging his book. The new media have no interest in mathematical advances. The math coming out of string theory is hard to explain. To sell physics research, it has to be a new discovery, or it has to be an Einstein essay. He could not get this essay published, if he removed the Einstein nonsense.

Medical Science recognizes Race and Sex

SciAm podcast: In recognition of Prostate Cancer Awareness Month this September, host Rachel Feltman sits down with Alfred Winkler, chief of urology at NewYork-Presbyterian Lower Manhattan Hospital, to discuss proactive steps individuals can take to protect themselves against prostate cancer. Black American men, in particular, face some of the highest rates of the disease in the world, with multiple factors contributing to this elevated risk. This episode explores efforts to raise awareness and promote early screening within this high-risk group. Other articles from the same magazine say Race Is a Social Construct and Human Sex Is Not Binary.

It is funny how our scientific institutions have been taken over by people willing to say obvious wrong things in order to serve misguided ideas about social justice.

Update: A new SciAm podcast today interviews a drag queen about teaching math.

SciAm goes Political Again; gets Trump Derangement

announces:

In the November election, the U.S. faces two futures. In one, the new president offers the country better prospects, relying on science, solid evidence and the willingness to learn from experience. She pushes policies that boost good jobs nationwide by embracing technology and clean energy. She supports education, public health and reproductive rights. She treats the climate crisis as the emergency it is and seeks to mitigate its catastrophic storms, fires and droughts.

In the other future, the new president endangers public health and safety and rejects evidence, preferring instead nonsensical conspiracy fantasies. ...

Only one of these futures will improve the fate of this country and the world. That is why, for only the second time in our magazine’s 179-year history, the editors of Scientific American are endorsing a candidate for president.

Harris and Walz are probably the most incompetent candidates for President in a long time. Harris avoids questions, and in the recent debate and interviews, she just babbled nonsense instead of answering the questions. When asked the simplest questions about what she plans to do or how she differs from Pres. Biden, she cannot answer.

This is science, so I looked for some example of Harris relying on science, following evidence, or learning from experience. No, there are no examples.

Trump touted his pandemic efforts during his first debate with Harris, but in 2020 he encouraged resistance to basic public health measures, spread misinformation about treatments and suggested injections of bleach could cure the disease. By the end of that year about 350,000 people in the U.S. had died of COVID; the current national total is well over a million.

This is all nonsense. Yes, thousands died under Trump, but more died under Biden-Harris. Trump never suggested injections of bleach. I don't think he even used the word "bleach". He merely commented on a scientific presentation on using disinfectants on surfaces to kill covid. Yes, Trump did downplay the shutdowns that health authorities were recommending, but we now know that the shutdowns did more harm than good.

It is hard to get any firm stand by Harris on any issue, except that she is pro-abortion, and against any restrictions on it.

Trump appointed the conservative U.S. Supreme Court justices who overturned Roe v. Wade, removing the constitutional right to a basic health-care procedure. He spreads misinformation about abortion—during the September debate, he said some states support abortion into the ninth month and beyond, calling it “execution after birth.” No state allows this.

Some states do allow abortions in the ninth month, and Trump correctly cited the Virginia governor as favoring it after birth.

The Biden-Harris administration’s 2023 Executive Order on Safe, Secure and Trustworthy Development and Use of Artificial Intelligence requires that AI-based products be safe for consumers and national security.

Really? Has that order made AI safe? Of course not. A science magazine should be able to back up what it says.

People can have political disagreements, and many will prefer Harris for various reasons. But a science magazine should stick to science. If it comments on a political race, it should be able to give evidence that a candidate is better for science.

There is no sign Harris understands any science issue. Reporters confronted her on her flip-flops on fracking, and she was unable to explain her views, and may not even know what fracking is.

New Research in Quantum Fault-tolerance

Dr. Quantum Computing has spent the last couple of years working for OpenAI, and returns to announce:

Let me end by sticking my neck out. If hardware progress continues at the rate we’ve seen for the past year or two, then I find it hard to understand why we won’t have useful fault-tolerant QCs within the next decade. (And now to retreat my neck a bit: the “if” clause in that sentence is important and non-removable!)

I think they created one logical qubit that lasts for a millisecond or so. Or something like that, I did not read the details.

I am just passing this prediction along. Seems doubtful to me.

The Invention of Large Language Models

Most people think that LLMs were invented by Google in 2016, or by OpenAI several years later.

AI expert Andrej Karpathy wrote on his blog:

The Unreasonable Effectiveness of Recurrent Neural Networks

May 21, 2015

The concept of attention is the most interesting recent architectural innovation in neural networks.

He constructs some LLMs, and his results seem pitiful compared to what is done today, but I would say he has a proof of concept.

Google introduced the transformer in Attention Is All You Need, a 2017 paper. As you can see, attention was already a hot idea at the time.

I am not sure who should get credit for inventing LLM. The basic ideas of neural nets go back decades. They got a whole lot smart when gaming GPU chips became fast and widely available, and AI researchers figured out how to use them efficiently.