General Relativity: Reopening the Lecture Archive

I am reopening my archive of general relativity lectures for Master’s students in theoretical physics.

The archive contains 54 videos in English and Arabic. Most recordings are from 2022, with earlier material from 2021 and 2020. The recordings vary in quality and presentation style. I had withdrawn them intending to produce improved versions—in language, scientific exposition, and visual presentation. Since I have not been able to carry out that project, I have decided to make the existing lectures available again, hoping they will remain useful to students.

The course follows six main stages:

1. Relativistic foundations: special relativity, the Minkowski metric, proper time, Lorentz transformations, and the equivalence principle.
2. Differential geometry: manifolds, vectors, geodesics, and tensors.
3. Curvature: parallel transport, covariant derivatives, connections, the Riemann curvature tensor, and geodesic deviation.
4. Einstein’s equations: Ricci curvature, the stress-energy-momentum tensor, perfect fluids, the Newtonian limit, and the derivation of Einstein’s equations from the Hilbert–Einstein action.
5. Exercises and further topics: the geometry of the two-sphere, the cosmological principle, maximally symmetric spaces, the Robertson–Walker metric, the ADM Hamiltonian formulation, and a review of differential geometry.
6. Supplementary cosmology and astrophysics: cosmic scales, flatness, topology, the size of the universe, and stellar ages and masses.

Many topics have alternative recordings in Arabic and English, but the pairing is not uniform. Some subjects have only one recording, while others are covered through several complementary lectures. In particular, the Arabic and English recordings numbered “Einstein’s equations 4” and “Einstein’s equations 5” cover related but different material.

Students need not watch every alternative recording of a given topic. They can choose the language and presentation that suit them, while consulting complementary recordings for additional material.

The downloadable table of contents includes the course outline and direct clickable links to all 54 videos.

Download the PDF table of contents:

https://drive.google.com/file/d/16OAE_Ieb4QcmE8K4vzbghvHowUk3Ocqa/view?usp=sharing

Watch the complete playlist:

https://www.youtube.com/playlist?list=PLsHtEvuFFjBsmCTY_lwlQK1cJe_VPAYSY

From Theoretical Physics to Artificial Intelligence, Consciousness, and Qur’anic Inimitability

1. Theoretical Physics: Is It Worth Studying?

If we want to identify precisely what theoretical physics is, we encounter quantum field theory, built on quantum mechanics and special relativity, alongside general relativity.

These are its two foundational wings, or the two pillars that allow us to understand the fundamental structure of matter, interactions, and spacetime, as well as the structure of the universe and black holes, at the deepest levels.

Studying them is immensely enjoyable, but it takes many years. Together, they give the physicist access to an enormous part of what we know about nature—although this is, of course, far from understanding everything.

But is studying them really worthwhile? I have discussed this question at length before. If we mean the practical and professional rewards for someone who devotes their life to them, my answer is: no, not really, in my judgment.

2. Particle Physics and Cosmology

Then come particle physics, astroparticle phenomenology, and cosmology. The motivation for studying these fields is clear and understandable: to discover what matter consists of, how interactions work, and how the universe developed and is structured. Most students who turn to theoretical physics enter through this door. That was my own experience too. Anyone who wants to study these subjects seriously needs the tools of quantum field theory, together with general relativity, particularly for cosmology.

But are they worth studying? They are no less worthwhile than field theory and relativity, since they are closer to physical phenomena and less mathematically abstract. Nevertheless, I do not see prospects that match the enormous effort they, too, demand.

I also believe that particle physics is struggling to achieve new fundamental breakthroughs. As for cosmology, despite the progress and precision of observations, the passage from data to fundamental conclusions often remains tied to assumptions and models that, in my view, deserve more caution than some confident presentations suggest.

3. Philosophy of Physics and the Philosophy of Spacetime

There is another direction: the philosophy of physics. Anyone interested in what I have called “the new kalām,” or in the philosophy of science more generally, should study the philosophy of physics. In my view, its central concerns are the philosophy of spacetime and its implications, and the philosophy of quantum mechanics and its implications.

But is studying the philosophy of physics worthwhile? If we are talking about practical and professional rewards, I do not think so either. Yet anyone who wants to philosophise about these questions should approach them properly and study the physics they intend to discuss.

This is why I am so surprised by the limited interest in my lectures on the philosophy of spacetime, particularly here on Facebook. Today, I finished the video on Kant’s philosophy. With a more engaged audience, I would have presented the entire journey: what I call the pre-classical philosophy, with its three pillars—Aristotle, Saint Augustine, and Ibn Arabi; then the classical philosophy of Newton, Leibniz, and Kant, whose lectures are available on YouTube; and finally the post-classical philosophy, beginning with Nietzsche, followed by Heidegger and Sartre, and also including Jung, alongside the revolutions brought about by Einstein and Bohr in relativity and quantum mechanics.

I have completed a paper on this subject, which will appear soon in the philosophy archive. This is a project I began sharing with you in 2018, if you remember, years before the appearance of today’s AI tools. I mention this to emphasise that our interest in these questions, and our writing about them, began long before these tools arrived.

4. From Condensed Matter to Quantum Computing and Artificial Intelligence

Now we come to condensed matter physics. This is an important field, directly connected to quantum computing and to the devices on which artificial intelligence depends, particularly through nanophysics and new materials. Many of its problems can also be approached from within theoretical physics itself.

Is it worthwhile? Technologically, certainly. Today’s computing depends on materials physics and semiconductors, while further directions include superconductors, photonics, and other technologies. As for the fundamental questions that personally occupy me in physics, however, I am not sure this is where I want to devote my remaining efforts—even while recognising that condensed matter physics has itself produced profound fundamental ideas.

But theoretical physics gave us quantum mechanics, and quantum mechanics made quantum computing and quantum information possible. This is where I see the most important remaining opening for myself in theoretical physics. That is, of course, my own view, not a binding judgment on the future of the entire field or on anyone else’s choices.

This is why I am trying to connect my research in quantum gravity with my studies in quantum computing and quantum information. Some time ago, I put forward a proposal that I consider very promising. Completing even part of it would be deeply satisfying to me.

I have also given lectures on quantum computing, but they attracted even less interest than the lectures on the philosophy of physics. We speak enthusiastically about science, technology, and the future, then turn away when actual study, concentration, and perseverance are required. We cannot consistently complain about the absence of serious scientific content and then ignore it when it is available. These lectures rest on years of study and work; I had expected those interested in these subjects to show a greater willingness to make the effort to follow them.

Quantum computing is, ultimately, computing. And computing is the foundation and backbone of artificial intelligence today. I believe AI will be one of the principal forces shaping humanity’s future, and I count myself among the theoretical physicists who are enthusiastic about it.

I do not mean that AI must wait for quantum computing to advance; it is already advancing through classical computing. What I mean is that quantum computing and quantum information offer me a natural path from fundamental physics to questions of information, learning, and intelligence.

5. Consciousness and Qur’anic Inimitability: A Testable Model

Some time ago, for example, I put forward what I consider an unconventional and highly ambitious proposal to study Qur’anic inimitability using artificial intelligence.

In this proposal, I draw a comparison between consciousness and the Qur’an. In the philosophy of consciousness, we encounter the “hard problem of consciousness,” connected to Thomas Nagel’s question: “What is it like to be a bat?” The issue is the subjective character of experience: what is it like for the creature having that experience? We can describe its processes and simulate its behaviour, but this does not, by itself, give us its personal experience from within.

Drawing on this comparison, I formulated the question of Qur’anic inimitability as follows: “What is it like for a language to be Qur’anic?” Within this conception, I assume that the essential nature of Qur’anic Arabic is accessible only to the Prophet, the angel, or God Almighty, with the meaning and level of that understanding differing in each case.

According to the model I propose, we can imitate ordinary Arabic texts and approach them without encountering an intrinsic minimum distance. Between the Qur’anic text and any attempt to imitate it, however, a minimum distance remains that cannot be crossed. It is as though the Qur’anic text lies at the centre of a ball in the space of Arabic language that imitation cannot enter.

This distance is more than a figure of speech. In the paper, I defined numerical measures of distance between texts and conducted preliminary experiments using lexical, semantic, logical, and stylistic measures. The hypothesis I want to test is that this minimum distance is an intrinsic property of the Qur’anic text, rather than merely a consequence of limitations in our tools of imitation and measurement. In this sense, I propose a Planck-like bound in language space: imitation improves, but a region remains around the text that it cannot enter.

There is, however, a fundamental difference between consciousness and the Qur’an. The gap associated with subjective experience applies to every conscious subject: simulating a person does not give us that person’s experience from within. The minimum distance hypothesised in this model, by contrast, applies specifically to the Qur’anic text, not to Arabic as a language or to every text written in it.

This is where AI becomes a means of testing the proposal: generating attempts at imitation, measuring their distances from the Qur’anic text, and comparing these with what we find for other texts. My preliminary experiments on English translations were intended to test the methodological tools. Testing the hypothesis itself requires working with the Arabic text.

If AI succeeds in imitating the Qur’an in a way that rules out this minimum distance, according to the specified measures and criteria, that would mean my proposed model does not capture the intended inimitability. It would not necessarily mean that inimitability itself had been disproved. What we are testing here is one particular model for understanding it.

6. Artificial Intelligence and Mathematical Physics: Three Axes

As I teach mathematical physics, I am now preparing a proposal entitled “Artificial Intelligence and Mathematical Physics,” organised around three axes and ten simulations.

The first uses AI as an extension of computational physics: we supply the physical model, and it helps us solve its equations and calculate its consequences.

The second uses AI as an extension of the theoretical physicist: we supply observations, and it helps us construct a description of the system by inferring its evolution law, discovering its equation, or proposing a correction to a known law.

The third studies AI itself as a physical system: we use mathematical physics and statistical mechanics to model networks and their learning processes, and to understand their behaviour.

7. Understanding Artificial Intelligence and a Criterion for the Emergence of Consciousness

I therefore believe that an important part of theoretical physics’ future passes through quantum computing and quantum information, and that this path opens a door towards artificial intelligence, in one way or another.

Just as theoretical physics, through the philosophy of spacetime, has helped us as theoretical physicists understand Kant’s immensely complex philosophy, I expect it to help us understand the physical and mathematical foundations of artificial intelligence.

For me, this goes much deeper than simply using AI as a tool in computational or theoretical physics. We want to simulate and understand AI itself using the tools of theoretical physics, particularly quantum mechanics and statistical mechanics, together with computational physics and quantum computing.

The question, then, is whether we can do this. Can we simulate artificial intelligence? Certainly, as far as its computational operations, learning dynamics, and behaviour are concerned. But I propose that the appearance of an intrinsic gap that simulation cannot cross would constitute a criterion for the emergence of artificial consciousness—provided that this gap belongs to the system itself, rather than arising from limitations in our tools or computational resources.

At that point, the hard problem of artificial consciousness takes shape: we can simulate the system’s operations and behaviour, but we cannot reach its subjective experience through simulation alone. The question becomes: “What is it like to be that system?”, just as Nagel asked: “What is it like to be a bat?”

Within my hypothesis, this intrinsic gap constitutes a criterion for the emergence of a conscious artificial subject whose behaviour can be simulated without possessing its inner experience. Here, the same structure appears that I proposed for human consciousness and hypothesised, in an analogous form, in my model of Qur’anic inimitability: simulation is possible, but an intrinsic minimum distance remains that it cannot overcome. The research challenge is to define and measure this gap, and to distinguish it from the limitations of our simulation tools.

What ultimately matters to me is that theoretical physics should become a path towards understanding AI itself, just as it has been a path towards understanding quantum computing and the philosophy of physics. For me, this is far more significant than simply using these fields as research tools and ways of earning a living in theoretical physics.

Quantum Field Theory: Reopening the Lecture Archive

 

I am reopening my archive of quantum field theory lectures for first-year Master’s students in theoretical physics taking QFT I and QFT II.

The archive contains 53 videos organized into 20 units. It brings together recordings from 2018 in mixed Arabic and English with recordings from 2021 in English and Arabic separately. The recordings vary in quality and presentation style. I had withdrawn them intending to produce improved versions—in language, scientific exposition, and visual presentation. Since I have not been able to carry out that project, I have decided to make the existing lectures available again, hoping they will remain useful to students.

The course follows five main stages:

1. Foundations: special relativity, electromagnetism, Hamilton’s principle, Lagrangian and Hamiltonian mechanics, and elementary particle physics.

2. The Dirac equation and spacetime symmetries: free solutions, rotations, spinors, and representations of the rotation and Lorentz groups.

3. Quantization of free fields: canonical quantization of scalar and Dirac fields, and their Feynman propagators.

4. Scattering and perturbation theory: the scattering matrix, Wick’s theorem, LSZ reduction formulas, and Feynman rules and diagrams for scalar phi-four theory.

5. Introduction to conformal field theory: the stress-energy-momentum tensor, the operator product expansion, primary operators, and the conformal anomaly.

Not every unit contains three alternative recordings. Eight units have two videos, eleven have three, and the final unit contains four successive lectures on conformal field theory. The 2018 recording on the scattering matrix and Wick’s theorem covers two units, while the unit on free solutions and symmetries of the Dirac equation includes an additional Arabic recording from 2021.

Students need not watch every version of a given unit. The alternative recordings generally cover overlapping material, allowing students to choose the language and presentation that suit them.

The downloadable table of contents includes the course outline, the recording exceptions, and direct clickable links to all 53 videos.

Download the PDF table of contents:

https://drive.google.com/file/d/1LfwPXHJ0_P6TACZXVj2LACbEvnHoGF_6/view?usp=sharing

Watch the complete playlist: 

https://www.youtube.com/watch?v=Y2bc6k1A3oM&list=PLsHtEvuFFjBswOlcNxWB7Os6g5FdLxH3p

 

When Rejection Becomes a Way of Life: The Struggling Actor as a Metaphor

What does it mean to be a “struggling actor”? Usually, it means someone trying to establish a career but unable to obtain enough paid work, recognition, or stability. The expression describes circumstances. It does not establish a lack of talent.

Is this merely an early phase? It can be. But it can also last decades, or an entire lifetime. There is no guaranteed progression from struggling to established. Some actors never secure a dependable livelihood through acting.

Someone may even begin with enormous promise, obtain encouraging early work, and then watch their career deteriorate. They can bear responsibility for decisions that contributed to this decline. But the outcome alone does not tell us how much was their responsibility. Others may never have received a genuine opportunity to demonstrate what they could do. A few small roles do not necessarily amount to a real chance.

Then comes the harder question: why continue after thirty years, perhaps into one’s fifties?

There are understandable answers. They love acting. They believe their abilities have matured. They still hope that one substantial role will allow their potential to become visible. Their ambition may be modest: steady work, a decent income, some recognition.

But what if the pursuit no longer sustains them? What if it brings sustained misery, financial insecurity, and an endless sequence of rejections? At that point, invoking love or a dream becomes insufficient. How much of a life should be surrendered to an opportunity that may never arrive?

Here, circumstances matter. An Algerian whose employment and mobility are severely restricted may be trapped in ways an American with accessible alternatives is not. American citizenship does not make every possibility available, but neither should substantial differences in freedom be dismissed. Where someone can realistically find another livelihood, pursue another direction, or move elsewhere, continued misery demands a stronger explanation than devotion to a vocation.

Perhaps they fear admitting loss. But after decades of disappointment, loss is already a lived experience. The more immediate fear may be the next rejection: another audition, another wait, another refusal. Repeated rejection can eventually become harder to bear than relinquishing the dream.

If viable alternatives exist, why should taking one count as failure? One need not conclude that the talent was imaginary, the effort worthless, or the treatment fair. It is enough to conclude that this particular path has become intolerable.

Someone can preserve their ability, and even continue their work, while withdrawing their livelihood and self-worth from the judgment of those who repeatedly refuse them.

The struggling actor is a metaphor for anyone whose promise remains unfulfilled. We should neither blame such a person automatically nor glorify their suffering. When opportunities are absent, persistence may reflect necessity. When alternatives are genuinely available and the pursuit brings only misery, changing direction becomes reasonable.

No dream acquires a permanent claim over a human life merely because that life has already given it thirty years.

If Authors Can Use AI, Why Can’t Referees?

 

The Reviewer Bottleneck  

Peer review has a very basic problem that predates artificial intelligence.

Reviewing a scientific paper properly is difficult work. It is also time-consuming work. A serious referee is expected to understand the central argument of a manuscript, check its important derivations, identify hidden assumptions, compare its claims with the existing literature, evaluate its novelty and significance, distinguish genuine errors from matters of presentation or interpretation, and determine whether the conclusions actually follow from the results. That is not a small task.

For a substantial theoretical paper, a careful review can consume many hours and sometimes several days of concentrated work. And who is supposed to do this work?

The same researchers who are already trying to write their own papers, conduct research, supervise students, teach courses, attend meetings, write reports and proposals, perform administrative duties, and maintain the rest of their academic lives.

Peer review depends overwhelmingly on scientists volunteering additional intellectual labor on top of everything else they already do.

It is therefore hardly surprising that journals increasingly struggle to find referees. Editors may have to invite many researchers before two agree to review a manuscript. Reviews can take weeks or months to arrive. Authors wait while editors search for another referee after several invitations have been declined. Sometimes the scientific evaluation itself takes much less time than simply finding someone willing and able to perform it.

This is not primarily a problem of bad referees. It is a problem of a system demanding substantial amounts of difficult intellectual work from people who often simply do not have enough time to perform it properly.

And this is precisely where artificial intelligence should enter the discussion.

AI Is More Urgently Needed by Reviewers Than by Authors

There is a curious asymmetry emerging in scientific publishing.

Authors are increasingly being given considerable latitude to use AI in their work, provided that they remain responsible for the final scientific content. They may use AI to improve language, organize arguments, summarize material, assist with coding, explore calculations, search literature, or interrogate complicated technical questions.

That development is understandable. But if AI assistance is useful to an author, it is arguably even more urgently useful to a reviewer.

The author is working on his or her own research. The author normally knows the calculation, the literature and the conceptual structure of the paper intimately. Months or years may have gone into producing it.

The referee enters from outside. The referee must understand somebody else's work, often under severe time constraints, reconstruct its logic, identify possible weaknesses, place it in the literature, and form a defensible scientific judgment.

This is exactly the kind of task for which intellectual assistance is valuable. Yet we are developing a publishing culture in which substantial AI assistance to the author is increasingly accepted while substantial AI assistance to the referee is often prohibited.

The priorities seem almost reversed.

If there is one part of the publication process where reducing the intellectual burden could immediately help the entire scientific system, it is peer review.

The Reviewer Does Not Need to Be Replaced

The issue is often framed incorrectly.

The choice is not

\[
\text{human reviewer}
\]

versus

\[
\text{AI reviewer}.
\]

The obvious model is

\[
\boxed{\text{human reviewer}+\text{AI assistance}.}
\]

The human referee remains responsible for the report.

That is exactly the principle already being adopted for authors. An author can use sophisticated computational tools without transferring authorship to those tools. The author remains responsible for the calculations and claims contained in the paper.

The same principle should apply to the referee.

Suppose I receive a theoretical-physics manuscript. Why should I not be allowed to give it to an AI system and ask:

“Reconstruct the central argument of this paper. Identify its principal assumptions. Check the important derivations where possible. Find possible inconsistencies. Compare the novelty claims with the cited literature. Separate potentially serious problems from minor ones.”

That does not replace me as referee. It gives me an extremely useful first analysis of the manuscript. I can then interrogate that analysis.

“Your objection to equation 27 seems incorrect because the large-(N) limit is taken before the infrared limit. Reanalyse it.”

“Derive equation 42 independently.”

“Where exactly is this assumption introduced?”

“Does reference 18 really establish what the authors claim?”

“Search for earlier work containing an equivalent result.”

“Construct the strongest argument against the main conclusion.”

“Now construct the strongest defense of the authors against those objections.”

At the end of this process, I decide what is correct. I decide what matters. I write the report. And I take responsibility for it.

This Could Make Reviews Better, Not Merely Faster

The obvious advantage is efficiency. A reviewer could perform in a few hours an analysis that might otherwise consume an entire weekend. Given the shortage of referees, that alone is important.

But speed is not the only issue. AI could also make reviews more thorough. A human referee may overlook an equation. A human referee may forget a relevant paper. A human referee may fail to notice that an assumption made on page 6 is incompatible with a conclusion on page 24. A human referee may simply become tired.

A sufficiently capable AI system can repeatedly traverse the manuscript, compare distant sections, track notation, examine limiting cases, summarize related literature and flag possible logical discontinuities.

Of course it can also make mistakes. Sometimes very convincing mistakes. But human referees make mistakes as well.

The correct response to this fact is not to prohibit assistance. It is to require the scientist to verify the criticisms that finally enter the referee report.

This is the same epistemic responsibility that scientists already exercise when using every other sophisticated tool.

We Do Not Demand That Referees Work Without Tools

Consider how strange such a principle would appear elsewhere. If an author evaluates a difficult integral using Mathematica, we do not insist that the referee reproduce the calculation by hand. If an author searches INSPIRE or Google Scholar, we do not require the referee to rely exclusively on papers remembered from years of reading. If an author performs a numerical computation on a powerful computer, nobody argues that the referee must reproduce it without computational assistance.

Scientific work has always advanced partly by developing better intellectual tools. AI is another such tool, although an unusually powerful one.

The relevant distinction should therefore not be

\[
\text{human work}\quad\text{versus}\quad\text{AI-assisted work}.
\]

It should be

\[
\text{responsible work}\quad\text{versus}\quad\text{irresponsible work}.
\]

Confidentiality Is a Solvable Problem

The strongest practical objection is confidentiality. An unpublished manuscript has been entrusted to the journal and to the referee. Clearly, a reviewer should not be free to distribute it indiscriminately to external parties.

But this is a technical and contractual problem. It is not a fundamental argument against AI-assisted reviewing.

Publishers could provide their own secure AI systems. A journal submission platform could make an AI assistant available directly to the referee. The manuscript would remain inside the publisher's protected infrastructure. The system could operate under explicit conditions that the manuscript is not used for training, is not retained beyond the required period, and is not disclosed externally. Universities could provide equivalent secure systems.

If confidentiality is genuinely the problem, then the rational response is to build confidential AI tools for reviewers. It is not to prevent reviewers from benefiting from AI.

Journals Should Give Every Referee an AI Assistant

This seems to me the natural direction in which peer review should develop. When a researcher accepts a review invitation, the journal system should provide an integrated AI assistant already authorized to examine the manuscript. The referee could ask:

What are the logically independent assumptions? 

Which conclusions depend on each assumption?

Check the derivation from equations 16 to 22.

Does the claimed limiting behaviour actually follow?

Which previous papers are closest to this result?

Are there important missing references?

Where does the manuscript move from demonstrated result to conjecture?

Construct the strongest possible objection to the central result.

Now defend the authors against that objection.

Such interaction would allow the referee to concentrate on the genuinely difficult part of scientific evaluation: understanding what matters. That is where human scientific experience remains indispensable.

The Referee Must Remain Responsible

There should nevertheless be a very clear boundary. A referee should not ask an AI system for a report and simply send that report to the editor. If the AI claims that an equation is wrong, the referee must satisfy himself or herself that the objection is valid before making it. If the AI claims that a result already exists, the referee should examine the relevant literature. If the AI misunderstands an unconventional argument, the referee must recognize the misunderstanding. The human reviewer remains scientifically and ethically responsible for the submitted report.

The governing principle is therefore simple:

\[
\boxed{\text{Do not outsource responsibility to AI. Outsource work to AI.}}
\]

This principle works equally well for authors, referees and editors.

AI Could Also Review the Reviewers

There is an additional possibility. AI could help improve not only manuscripts but referee reports themselves. Before sending a report to the author, the editorial system could ask:

Does this criticism accurately represent what the manuscript says?

Is the referee asking for work far outside the stated scope of the paper?

Does the claim that the result is already known have supporting references?

Are requested citations actually relevant?

Does the referee appear to have misunderstood a definition or convention used by the authors?

This would introduce some quality control into a part of scientific publishing that currently receives remarkably little.

Papers are scrutinized intensely. Authors are scrutinized intensely. Referee reports themselves are much less systematically scrutinized. AI could help correct that imbalance. 

A further problem is the imbalance of power built into the present system. A referee can sometimes become effectively dictatorial: demanding additional calculations, imposing personal preferences about scope or interpretation, insisting on unnecessary revisions, or repeatedly moving the conditions for acceptance, while facing very little accountability for those demands. 

Editors, especially when overloaded or working outside the referee's exact specialization, may simply defer to the report rather than independently judging whether the requests are scientifically justified. The result is an asymmetric process in which authors must answer every objection in detail, while referees are rarely required to justify the proportionality, relevance, or even correctness of their criticisms. 

AI-assisted editorial checking could provide a useful counterweight by asking whether each major referee demand is actually supported by the manuscript, relevant to its claims, and necessary for establishing the validity of the work.

The Strange Reversal of Priorities

This brings us back to the central paradox. Scientific publishing is suffering from a shortage of reviewers. Careful reviewing requires substantial time and intellectual effort. Researchers are increasingly reluctant or simply unable to accept additional referee assignments. Editors struggle to obtain reports. Authors wait. And yet, when one of the most powerful tools ever developed for reducing intellectual workload becomes available, we give considerable latitude to authors while being much more restrictive with reviewers. This seems backwards.

Authors certainly benefit from AI. But the publication system may need AI-assisted refereeing even more urgently.  The appropriate future of peer review is therefore not the replacement of scientists by machines. It is

\[
\boxed{
\text{AI assistance}
+
\text{human scientific judgment}
+
\text{human responsibility}.
}
\]

Used in this way, AI would not weaken peer review. It could help rescue a system that is already struggling under the amount of work we ask human referees to perform.