Juan Maldacena: The Emergence of Spacetime

ELI5/TLDR

Juan Maldacena is the physicist who, as a graduate student in 1997, wrote what’s now the most-cited paper in theoretical physics. His big idea — usually called the “AdS/CFT correspondence” or “holographic principle” — says that gravity inside a region of spacetime can be exactly described by a quantum theory living on the boundary of that region, like a 3D scene encoded on a 2D surface. In this conversation he walks through where that idea has gone in 30 years: that spacetime might not be fundamental at all but stitched together out of quantum entanglement, that wormholes are “leaky pipes” he’s still trying to understand, that the inside of a black hole is “a place in the future” rather than a location, and that you cannot do physics from a god’s-eye view — every observer is part of the system. He hedges everywhere, which is the honest move. Stay for the end: the man who wrote the most-cited paper in theoretical physics says he didn’t feel good enough as a graduate student.

The Full Story

What is spacetime made of?

Curt opens with the bluntest possible question. Maldacena’s answer is precise and surprising:

“Spacetime in the theory of general relativity, it’s not made out of anything. It’s a primary concept. It’s the main dynamical object of the theory.”

In Einstein’s classical theory, spacetime is the stage. You don’t ask what the stage is made of. The question only becomes meaningful when you try to write a deeper theory underneath general relativity — a quantum theory of gravity. And there, Maldacena says, the most useful answer of the last twenty years has been: spacetime might be built out of qubits, basic quantum bits of information, that live on a faraway boundary surrounding the region of spacetime in question.

Imagine a glass aquarium. The water inside is the spacetime — its curves, its black holes, its galaxies. Now imagine that everything happening in the water can be perfectly described by some intricate pattern of vibrations on the glass walls. The glass is the boundary; the water is “the bulk.” The astonishing claim is that the wall pattern contains every last fact about the water. That’s holography in one paragraph.

This is what physicists mean by an emergent description: the bulk spacetime is not fundamental, it “emerges” from the boundary degrees of freedom — the qubits on the glass.

Why combining gravity and quantum mechanics is hard

The pop-science answer is “one is continuous, one is discrete.” Maldacena pushes back gently:

“In quantum mechanics there is usually some time, some order between operators… in general relativity, spacetime can have different geometries, different topologies — we don’t know what the order is. Also, in quantum mechanics we have some observer who’s outside the system, and in gravity everything is somehow inside the system.”

Two problems, both conceptual. First: quantum mechanics is built around the idea that you measure things in a particular order, and “before” and “after” mean something fixed. Gravity allows spacetime itself to fluctuate — different shapes with different notions of time, all superposed. Whose “before” wins?

Second, and deeper: every textbook quantum mechanics problem has an observer outside the system, looking in with no mass or energy, just recording outcomes. There is no such observer for the universe. Anyone measuring anything is themselves a piece of the universe, with their own mass, energy, gravitational pull. The “view from nowhere” doesn’t exist. Maldacena will return to this several times.

Black hole interiors and the meaning of “singularity”

In a black hole, Einstein’s equations predict that spacetime curvature becomes infinite somewhere inside. That somewhere is called a singularity. Maldacena makes a surprising move on this:

“The singularity is not a place inside the black hole. It’s a place in the future.”

If you fell into a black hole, you wouldn’t reach the singularity by traveling further inward in space. You’d reach it by waiting. Once past the horizon, the singularity is in your future the way next Tuesday is in yours: unavoidable, no matter which direction you face. He also reframes what “singularity” means as a word:

“Singularity is just the name for things we don’t understand.”

In other words: it’s a label for the place our equations break. A flag in the ground saying “theory ends here, please send a better theory.” The thing that needs fixing is not the black hole. It’s our description of it.

He compares the inside of a black hole to a tiny “big crunch.” Our overall universe is expanding — a big bang in our past, ongoing expansion. But in regions where matter is dense enough, gravity reverses the expansion locally and matter collapses. A black hole interior is a small region of universe collapsing to infinite density. Same equations, mirror direction.

The Ryu-Takayanagi formula: where geometry meets information

Around 2006, two Japanese physicists, Shinsei Ryu and Tadashi Takayanagi, found a strange formula. To compute a particular kind of “fine-grained” entropy of a quantum system on the boundary, you reach into the bulk spacetime, find a specific surface inside it, and measure that surface’s area. The area of a geometric surface tells you how much quantum information is in the system.

This was a clue, not a closed case. It suggested that geometry and quantum information are deeply, mathematically the same kind of thing. Entropy — in quantum mechanics, the amount of information you’d need to fully describe a system — has units of bits. Area has units of square meters. Yet here was a formula equating them, modulo Planck’s constant. Maldacena puts it carefully:

“[This] represents a connection between space-time geometry and quantum information.”

Bridge: think of entropy here not as “disorder” (the word Shantum may have learned in chemistry class) but as “amount of information.” A perfectly ordered crystal has low entropy because you can describe it in a sentence. A scrambled mess has high entropy because it takes many bits to specify. The Ryu-Takayanagi formula says: the amount of information stored in a chunk of the boundary equals the area of a particular surface in the bulk.

Entanglement: the quiet superpower

Quantum entanglement is the phenomenon Einstein, Podolsky, and Rosen identified in 1935 — usually shorthanded as EPR. Two particles, prepared together, can end up in a state where measuring one instantly tells you something about the other, no matter how far apart they are. Not magic, not faster-than-light signaling — just a correlation that has no classical analog.

In the same year, 1935, Einstein and Rosen wrote a separate paper describing what looks like a wormhole — two black holes joined through their interiors by a bridge. Decades later, a physicist named Kruskal showed they were exactly that: two universes connected by a hidden tunnel. This is called the Einstein-Rosen bridge — ER for short.

Maldacena’s now-famous slogan, written in an email to Leonard Susskind, was “ER = EPR.” Six characters. The wormhole equals entanglement. Or, more carefully: two black holes that are entangled in a particular way (the “thermofield double” state, if you must) give rise to a geometric wormhole connecting them. The geometric bridge and the quantum correlation are the same fact, viewed two ways.

“If they’re entangled in this particular entangle state, then the idea is that they give rise to a geometry, connected geometry… In this particular example I think the arguments that the entanglement creates this space-time geometry, this space-time connection, is fairly clear and quite convincing.”

But — and Maldacena is careful here — the wider claim that all entanglement creates all spacetime geometry is more aspirational. If you entangle two particles in a lab, no actual wormhole appears that you can drive a truck through. The Einstein-style geometry isn’t there. ER = EPR is a slogan, a principle, a north star — a guess at what a deeper theory should look like, in which some generalized notion of geometry becomes real even for two entangled spins.

Wormholes: the leaky pipes

The metaphor that opens the episode:

“Wormholes are a bit like leaky pipes… Not everything is fitting together.”

Two kinds matter. The first is what we just described — Einstein-Rosen bridges, formed automatically when two black holes are properly entangled. These are non-traversable: you cannot send signals through them. This is consistent. Entanglement alone never lets you send signals.

The second is “traversable wormholes,” which Maldacena and collaborators have shown are mathematically possible if you let two black holes interact gently with each other. They form a geometric tunnel you could in principle pass through. Important caveats: you cannot use them to travel faster than light through the surrounding space — they’re more like long, deep tunnels under a mountain than the warp-drive shortcuts of Star Trek. They’re also classically forbidden — they exist only thanks to quantum effects, where energy can briefly go negative.

His analogy: imagine digging a frictionless tunnel through the center of the Earth. From the outside, two cities are far apart. But if you slid down the tunnel, you’d reach the other side in 40 minutes. The traversable wormholes Maldacena studies are spacetime versions of this: long detours that, from the inside, feel short.

“These are not things that will exist in nature. They would require… new particles that we haven’t seen.”

Solutions, not facts. He’s careful.

The lab “wormhole” headlines

A few years ago, headlines claimed scientists had created a wormhole inside a quantum computer. Maldacena’s read: a small group of researchers ran a quantum simulation with seven qubits per side that displayed some features mathematically similar to a tiny wormhole. Whether this counts as “creating” or “simulating” a wormhole is, he says with patience, a philosophical question:

“It’s a bit like this question of when a sand pile becomes a sand pile. If you have seven grains of sand, okay, maybe it’s not a sand pile, but once you have enough grains, it will become more universally recognized as a sand pile.”

The headlines oversold it. The work itself is a real first step toward something genuinely strange.

The black hole information paradox and the page curve

If you throw an encyclopedia into a black hole and it slowly evaporates as Hawking radiation, where did the information in the encyclopedia go? Stephen Hawking originally argued the radiation is featureless — pure thermal noise, no fingerprints — which would mean information was lost. That breaks quantum mechanics, which is built on information being preserved. (The technical word for this preservation is “unitarity.”)

Don Page proposed a curve — the page curve — describing how information should flow back out if the black hole behaves quantum-mechanically. Information rises, peaks, then falls back to zero as the black hole fully evaporates. For decades, deriving the page curve from gravity was beyond reach. The recent breakthrough — the “island formula” that Maldacena and several collaborators developed — finally lets physicists compute it directly. Information, it now appears, does come out. Hawking was wrong, but his question was the right one.

The shredded-document analogy he uses with Curt:

“It’s like in detective shows when they try to take the shredded documents and put it together. It’s a more complicated version of that.”

Burn paper, mix the ash with the radiation in the room, and in principle the information is recoverable. The black hole version says: same is true here, but only if you understand the geometry of the interior properly.

Why every observer matters

Recent work has formalized the role of the observer. The technical version uses heavy mathematics (“type two and type three algebras,” for the curious). The conceptual version is: in quantum gravity, you cannot leave the observer out. There is no view from nowhere. The clock you use to measure time, the ruler you use to measure distance — these are physical systems with mass and energy, and they participate in the gravity they’re measuring.

Maldacena’s paper “Real Observers Solve Imaginary Problems” plays on this. There’s a calculation in de Sitter space (the geometry of an expanding universe like ours) where the count of quantum states comes out to an imaginary number. Negative or complex. Physically nonsensical. Add a real observer to the calculation — a thing in the universe that has a clock and a position — and the imaginary number becomes a positive integer. The observer is not a nuisance to be abstracted away. The observer is what makes the answer real.

“Can you measure time without the clock? And the idea is that you can’t. You would need a clock of some kind to be able to talk about time.”

Doing physics with AI

Curt asks if Maldacena uses AI in his work. His answer is unfussy: yes, mostly as a better Google search and an integral-checker. He’s not written a paper with AI. But Andy Strominger recently used GPT to discover a new formula for gluon scattering amplitudes, and Maldacena is now doing follow-up work with one of Strominger’s collaborators on the same formula. He sounds optimistic but humble:

“I would advise people not to imitate what I do in AI. I think maybe I feel already like a dinosaur that I’m not learning it fast enough."

"Spacetime is doomed”

Curt drops Nima Arkani-Hamed’s slogan. Maldacena traces it back: in 1908, Hermann Minkowski declared “space alone, and time alone, are doomed to fade into mere shadows” — what we have now is spacetime. A century later, physicists are saying spacetime itself will fade in favor of some deeper concept we haven’t named yet. Qubits on a boundary? Entanglement structures? Scattering amplitudes? Nobody knows.

“It’s a challenge for us to find the new concept.”

Advice and honesty

The most generous part of the conversation comes near the end. Maldacena describes feeling, as a graduate student, that he wasn’t good enough — that the great work had already been done, that his time had come too late, that he wouldn’t measure up. This from the man who would write the most-cited paper in theoretical physics during exactly that period.

“I think it’s common, maybe, not to feel good and to feel that maybe you’re not good enough… but eventually you’ll make your contributions.”

His advice for students is plain: be open-minded, question things, don’t repeat what everyone says, understand things your own way. His example is from his own work on cosmological perturbation theory — he found the literature confusing, went back to the harmonic oscillator, the simplest thing, redid the calculations from scratch. Then he could read everyone else’s papers and finally see what they meant. Lore in a field is dangerous. Calculations are not.

Key Takeaways

AdS/CFT correspondence (the Maldacena conjecture, 1997): a gravity theory in a region of negatively curved spacetime (“AdS”) is exactly equivalent to a quantum field theory living on the boundary of that region. Bulk gravity = boundary quantum mechanics. This is “holography” formalized.
Spacetime is not fundamental. In Einstein’s theory it’s the stage. In quantum gravity, it likely emerges from quantum information — qubits on a distant boundary.
Black hole singularities are “in your future,” not at a location inside. Once past the horizon, the singularity is when, not where. They mark the failure of current physics, not real physical points.
Ryu-Takayanagi formula (2006): the entropy of a region of the boundary equals the area of a specific minimal surface inside the bulk. Geometry is information, information is geometry.
ER = EPR: an Einstein-Rosen bridge (a wormhole) is the geometric face of EPR-style quantum entanglement. Two black holes correctly entangled give rise to a non-traversable wormhole between them. Whether this generalizes to all entanglement is open.
Traversable wormholes are mathematically possible thanks to quantum effects (negative energy), but require physics beyond the Standard Model. They wouldn’t beat the speed of light through ambient space, only offer long shortcuts.
The black hole information paradox is closer to resolved than it was a decade ago. The “island formula” lets physicists compute the page curve from gravity, showing information leaks back out as a black hole evaporates.
No view from nowhere. Quantum gravity requires the observer to be inside the system. “Real observers solve imaginary problems” — adding an observer with a clock turns nonsensical complex answers into real ones.
Wormholes (“leaky pipes”) are the open frontier. They are useful for deriving formulas but raise unresolved questions about whether the constants of nature should be averaged over or held fixed.
De Sitter / cosmology side is harder than AdS. AdS/CFT works because of negative curvature plus extra symmetries. The corresponding theory for our actual expanding universe (de Sitter) doesn’t yet have working examples — possibly because the duality is only approximate there, possibly because we haven’t found the right tools.

Claude’s Take

This is a near-ideal physics interview. Maldacena hedges where he should hedge — the slogan ER = EPR is “an aspiration,” the lab wormhole is “debatable” whether it counts as creation or simulation, the entropic gravity program “might be true” in some version. He distinguishes solutions (mathematically allowed) from realizations (actually present in nature). He distinguishes manifest locality (locality you can see in the equations) from physical locality (no faster-than-light signaling). These distinctions are exactly what gets crushed in pop-science treatments where every conjecture becomes a discovery.

The spine of his career is one bet: that gravity and quantum information are the same subject. From that bet has come AdS/CFT, ER=EPR, the entropy formulas, the recent island formula. The bet has paid off enough times that “spacetime is qubits on a boundary” is now a respectable working hypothesis, not a fringe view. But he’s frank that the cases where this fully works are unrealistic — negative curvature, supersymmetry, no expanding universe. The version that describes our actual cosmos is missing.

The closing material is unusually generous. A theorist of his stature, on camera, saying he didn’t feel good enough as a student, that his students often have ideas he initially thinks are wrong, that he’s a “dinosaur” with AI — this is how someone with nothing left to prove behaves. It’s worth the watch for that alone, even if the physics is too dense in places.

Score: 9. The half-point comes off only because Curt occasionally lobs jargon-heavy questions (“entanglement wedge reconstruction,” “JT gravity,” “type three to type two algebras”) that get partial answers. A viewer without grad-school physics will lose the thread on those. But the core of the conversation — what is spacetime, why quantum mechanics and gravity fight, what black holes really are — is some of the cleanest exposition of frontier theoretical physics you can find anywhere.