Published: June 04, 2026, 03:34 UTC
For half a century, physicists have chased a holy grail: a theory that marries general relativity — Einstein’s majestic description of gravity as warped space-time — with the counterintuitive rules of quantum mechanics. It is the single biggest open problem in fundamental physics, and progress has been slow. But a pair of recent papers, building on ideas from quantum information theory, suggest a strange new clue. The missing ingredient, some researchers now suspect, may be something called “quantum magic.”
Yes, that’s its real name.
The space-time fabric party
Here’s the picture that has emerged over the last thirty years. In the 1990s, theoretical physicist Juan Maldacena discovered a profound equivalence — a “duality” — between certain gravitational theories and quantum field theories living on a lower-dimensional boundary. The idea, known as the holographic principle, suggests that our three-dimensional universe might be a kind of holographic projection of information encoded on a distant, two-dimensional surface.
Think of it this way: everything that happens inside a volume of space can be fully described by data on the volume’s boundary. Gravity, in this picture, is not fundamental — it emerges from the quantum behavior of particles on that boundary.
In the mid-2000s, another insight arrived. Maldacena and Leonard Susskind proposed the ER=EPR conjecture: that entangled particles (the “spooky action at a distance” Einstein famously disliked) are connected by wormholes — tiny tunnels through space-time. Entanglement, in other words, literally builds the geometry of the universe. Where particles are entangled, space exists. Where they aren’t, there is only void.
This was a breathtaking leap. It suggested that space-time is not a stage on which quantum mechanics plays out. Space-time is quantum mechanics — specifically, a vast network of entangled qubits.
The stabilizer trap
If entanglement builds the structure of space, you might think the problem is solved. Just take a bunch of entangled quantum particles and — presto — out pops gravity, right?
Not quite. Over the last decade, researchers working on “holographic codes” — mathematical toy models of the holographic principle — discovered something surprising. Certain kinds of entangled states, called stabilizer states, could reproduce the geometry of space-time beautifully. These states, built using only Clifford gates (the quantum operations Hadamard, S, and CNOT), produce a kind of frozen, crystalline space. The shape is there, but nothing happens. Matter cannot interact with geometry. It’s like building a perfect, ornate stage and then realizing the actors are glued to the floor.
The reason? Stabilizer states can be efficiently simulated on a classical computer. This is the punchline of the Gottesman-Knill theorem, one of the most celebrated results in quantum computation: any quantum circuit built exclusively from Clifford gates can be run on a classical machine with polynomial resources. There is no computational surprise, no oomph. And in the holographic picture, that lack of oomph translates to inert, rigid space.
Enter the wizards.
What is quantum magic?
In quantum computing, “magic” has a very specific technical meaning. It refers to non-stabilizerness — the degree to which a quantum state or operation cannot be simulated efficiently on a classical computer. Magic is what gives quantum computers their power.
The fundamental gates of quantum computing — Hadamard, S, and CNOT — are “Clifford gates.” They do interesting things, but they are boring in a specific sense: a computer made only of Clifford gates offers no advantage over a classical machine. To reach true quantum advantage, you need something extra. You need a T gate (a π/8 phase rotation), or a Toffoli gate, or some other operation that breaks the stabilizer symmetry. These are the magic gates.
The amount of magic in a quantum state can be measured. Researchers have developed resource theories around it, with names like “mana” and “stabilizer entropy.” It sounds playful, but the mathematics is serious. Magic is the essential fuel for quantum computation.
The fabric softener of space
In 2025, a team led by researchers including Cao, Cheng, Hamma, Leone, Munizzi, and Oliviero published a paper in PRX Quantum that passed peer review. They proved a crisp, elegant result: in holographic codes, if the “non-local magic” vanishes, there is no gravitational backreaction. In plain English: if your quantum system has no magic, matter cannot bend space-time.
Now, in March 2026, the same team — joined by Karthikeyan, Li, and the legendary quantum information theorist John Preskill — released a massive 98-page paper on the arXiv (2603.13475, not yet peer-reviewed) that deepens the argument considerably.
The key idea is this: holographic codes built from stabilizer states produce empty, rigid space-time. They are perfect error-correcting codes — every piece of quantum information is perfectly protected from local errors. But a perfect code is also a static code. Nothing can change. There is no dynamics.
When you introduce non-local magic — by applying non-Clifford gates to the code — something remarkable happens. The perfect error correction becomes approximate. The neat separation between “inside” and “boundary” blurs. Information can leak between the two. And that leakage, the authors argue, is precisely gravitational backreaction. It is the mechanism by which matter perturbs geometry.
Cao described magic as “the fabric softener of space.” Without magic, space is stiff, rigid, unyielding. Magic loosens it, allowing matter to create dents and curves — allowing gravity to exist.
Preskill, in an interview with Quanta Magazine, put it more soberly: “Without magic, things are a little too simple. And, you know, quantum space-time isn’t quite that simple.”
Einstein’s apple
There is something deeply satisfying about this picture. The “apple” that Newton watched fall — the archetypal example of gravity at work — was, in this framework, a manifestation of imperfect quantum error correction.
Think about that. When you hold an apple and let it go, it falls because the quantum information describing that apple is not perfectly hidden from the geometry of space. There is a tiny bit of leakage between matter degrees of freedom and geometric degrees of freedom. That leakage, made possible by the presence of magic in the underlying quantum state, is what we experience as gravity.
Entanglement gives space its shape. Magic gives it its heft.
Step 0.5 of 5
It would be irresponsible to suggest that the problem of quantum gravity is solved. Cao himself calls this work “Step 0.5 of 5.” The models used so far describe anti-de Sitter space — a negatively curved universe that is not our own. They do not yet capture Einstein’s equations in full. They do not include time — a rather crucial omission. And the March 2026 paper has not yet passed peer review.
What the work does establish, rigorously, is a necessary condition: for gravity to emerge from a holographic code, magic must be present. The converse — whether magic is sufficient to produce gravity — remains an open question. But the evidence is mounting.
Why this matters beyond physics
The connection between magic and gravity has a striking practical implication. If simulating gravity requires quantum states that cannot be efficiently simulated classically, then understanding holographic gravity may intrinsically require a quantum computer. The problem may be, in the strict computational sense, hard — impossible to shortcut with clever classical algorithms.
This ties together two of the most exciting frontiers in modern science: the quest to understand quantum gravity, and the quest to build a useful quantum computer. They may be two sides of the same coin.
“We’re trying to understand the ultimate nature of reality,” Preskill said, “and it turns out that the key to doing that may be the same thing that makes quantum computers powerful.”
Magic, in other words, is not just a computational resource. It may be a cosmic one.
The main paper discussed is: Cao, Cheng, Karthikeyan, Li, Preskill, “Non-local Magic and Gravity” (arXiv:2603.13475, March 2026). Precursor: Cao, Cheng, Hamma, Leone, Munizzi, Oliviero, PRX Quantum 6, 040375 (2025). This article was informed by reporting by Charlie Wood in Quanta Magazine (June 3, 2026).