Has Quantum Supremacy been Achieved?

Dominik Hangleiter writes in a new paper:

Recently, I gave a couple of perspective talks on quantum advantage, one at the annual retreat of the CIQC and one at a recent KITP programme. I started off by polling the audience on who believed quantum advantage had been achieved. Just this one, simple question.

The audience was mostly experimental and theoretical physicists with a few CS theory folks sprinkled in. I was sure that these audiences would be overwhelmingly convinced of the successful demonstration of quantum advantage. After all, more than half a decade has passed since the first experimental claim [AAB+19] of “quantum supremacy” as John Preskill called the idea “to perform tasks with controlled quantum systems going beyond what can be achieved with ordinary digital computers” ...

I could not have been more wrong: In both talks, less than half of the people in the audience thought that quantum advantage had been achieved.

After several pages describing the experiments, he concludes:

I hope that I could convince you that quantum advantage has been achieved. There are some open loopholes, but if you are happy with physics-level experimental evidence, then you should be convinced that the RCS experiments of the past years have demonstrated quantum advantage.

No. The whole point of quantum supremacy, aka advantage, is to do an experiment that convincingly demonstrates that super-Turing computers are possible. If most of the experts have not been convinced, then the principle has not be demonstrated.

I am a skeptic, and will be hard to convince. But they have not even convinced the experts who work in the field.

But read the paper, and make up your own mind.

Scott Aaronson seems to be still on the fence. His recent postings have been either about how he hates Pres. Trump, or how he agrees with Trump's pro-Israel foreign policy.

Science Papers are now mainly read by AI LLMs

Alexander Kustov, with Claude AI assistance, writes Academics Need to Wake Up on AI and Part II:

1. AI can already do social science research better than most professors.

2. The academic paper is a dead format walking.

3. The commercial journal system may not survive this.

4. Academics hold AI to absurd double standards.

In particular: Most papers are already mostly read by AI, not humans. Your primary audience is increasingly LLMs.

Physics Lifetime: 1820 to 1970

Popular tweet:

Many people don't understand just how brutal diminishing returns in theoretical physics were.

Physics barely existed before 1820. After 1970, there was essentially nothing left to discover.

In 1819 there were probably less than 100 full-time paid physicists in the whole world.

By 2026 there are probably about a million physicists across academia and industry, and that number was already huge in the 1970s when physics sort of "ended" with QCD and electroweak unification.

A small, brave band of gentlemen-scholars and amateurs worked out the most important parts of physical law in the 1800s. People doing it as a hobby!

Today, vast armies of professionals equipped with supercomputers toil away in the quantum gravity dungeon, unable to make progress.

Diminishing returns are brutal.

my point is that the low hanging fruits of physics were all picked in a brief window from about 1820 to 1970.

Before that, it was difficult to get anything done at all, there was no funding

Quantum Computing and National Security

The Wash. Times reports:

One of the most advanced technologies intersecting with U.S. national security today is quantum computing. Quantum has arrived in 2026, and how it ultimately gets implemented will impact America’s standing in great power geopolitical competition, especially with U.S. adversaries. National Security Editor Guy Taylor sits down with industry leaders at “Qubits26 Quantum Realized,” a conference hosted by D-Wave Quantum, for a wide-ranging discussion on what quantum computing is and how it stands to change the world.

The rest of the article is paywalled, and I do not need to read it. It is all a scam. Quantum computers will not affect national security.

Good Wlll Hunting

SciAm reprints an article on Why mathematicians hate Good Will Hunting.

With the award ceremony for the Oscars this month, many people are thinking back on past winners—including Good Will Hunting. It’s worth taking a closer look at the blackboard in a film that, in 1997, took nine nominations and won for both original screenplay and actor in a supporting role. ...

But I still think the filmmakers chose this particular math problem poorly, even for a Hollywood film.

No, that is not why mathematicians hate the movie.

The hero is a fictional exceptionally talented math prodigy. Supposedly he enjoys math so much that he gets a job as an MIT janitor, and eavesdrop on the research there.

But he never spends any of his free time doing math. Instead he goes drinking with his non-math buddies, and getting into fights. In the end, he decides that math is for losers, and he abandons a wonderful math opportunity in favor of chasing a girlfriend.

Nobody gets that good at math unless he enjoys it very much. The movie fails to portray that at all.

Nature Deepseek Article may be Exaggerated

The 17 September 2025 Nature cover story was this article by China AI company Deepseek. Supposedly it had a hot new AI model trained at low cost.

They aRe widely accused of using pirated Nvidia chips that were illegally imported.

Now Anthropic claims that Deepseek and other China AI companies have been using thousands of bogus accounts to steal data from Anthropic models.

Deepseek has made some decent AI LLMs, but maybe not the scientific advance they claimed. Nature said that it investigated the validity of the Deepseek claims, but I don't know how it could know what was going on in China.

Anthropic could be lying. It really hates Chinese competition. So I am not necessarily drawing any conclusions here.

Why Chance make no sense in Many-Worlds

I have argued here that many-world (Everettian) theory is incoherent because it fails to make any sense out of probability. To believe in the theory, you have to reject probability as widely understood, and then to reject all of science.

No one admits this, of course, but see how a new paper dances around the subject.

Everettian chance in no uncertain terms

The current landscape of views on chance in Everettian quantum mechanics is a curious one. On the one side, longstanding critics of many-worlds theories maintain that probability is needed to make sense of the machinery that Everettians use to derive chance values, resulting in circularity (Baker 2007, Dawid and Thébault 2015, Mandolesi 2019). On the other side, Everettians seem to agree that chance values should be derived in terms of agents’ uncertain or partial beliefs—but they cannot agree on how.1

Perhaps the most famous of these uncertainty-based approaches is the decision-theoretic program explored by Deutsch (1999) and Wallace (2012): they purport to prove that a rational Everettian agent must order their preferences over acts in a way that recovers the Born rule. Sebens and Carroll (2018) take issue with one of the principles of rationality in Wallace’s approach and instead aim to derive Everettian probabilities from principles governing self-locating uncertainty. McQueen and Vaidman (2019) offer yet another self- locating uncertainty approach, taking issue with Sebens and Carroll’s metaphysical view of branching. Notably, all three approaches claim that symmetries of quantum states are central to their arguments, but none attempt to characterize the symmetries at play.2

The most common defense of many-worlds here is to give up on probabilities directly, but to argue that it could be rational to believe in them anywy, as some sort of subjective way of trying to make sense out of nonsense.

Any sort of scientific outlook requires saying that some things happen, some do not happen, and some things are more likely than others. Many-worlds is unable to deal with any of this. Just look at the above nonsense. It essentially says that many-worlds researchers are trying to make sense out of chance, and not succeeding.