· Nov 17, 2026
Anthropic says Claude solved two open problems in black hole physics
Markdown · JSON · Original source
reports that completed a long-standing calculation about whether a phase transition remains abrupt in a quantum theory, and established an exact limit for a technique used to rule out candidate theories of quantum gravity. The results concern theoretical model universes, not our own universe directly, and Anthropic says neither has practical applications. For the first problem, Anthropic says the transition is abrupt both in the regime described by Hawking and Page and in a regime where spacetime loses a definite shape. It says this suggests black-hole formation is built into the underlying quantum theory, while the behavior between those regimes remains unestablished. For the second, Claude used a spectral curve to determine the optimal ceiling for the simplest version of the modular bootstrap. The result narrows the search but does not identify the correct theory. Anthropic says agents worked on the problems for about six weeks, and that the modular bootstrap result was in and checked by a physicist. The two results involve different work: completing a difficult calculation and using a mathematical tool to establish a limit.
Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.
Original source
· Article
Source details
Original source date: Nov 17, 2026 (2026-11-17T00:00:00.000Z).
Date basis: Date stated by the publisher.
Added to Escape Meridian: 2026-11-18T20:02:40.772Z. First observed: 2026-11-18T18:04:05.367Z. Retrieved: 2026-11-18T19:47:01.288Z.
Referenced by
Recorded citations in public readings. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
describes theoretical model universes; it says neither result describes our own universe directly or has practical applications. The first result establishes the abrupt transition at the two studied extremes; Anthropic presents persistence between them as the most natural possibility, not as established. The modular bootstrap result concerns the simplest version of the method, which ignores how black holes spin, and does not identify the correct theory.
Open question
What further ingredients would be needed to push the modular bootstrap ceiling lower?
Source-stated
Anthropic reports that Claude completed a calculation finding the phase transition abrupt in a quantum theory, as well as in the regime described by Hawking and Page.
It showed that the phase transition is abrupt. The same sudden switch appears at both extremes
Source-stated
Anthropic says the result suggests black-hole formation is built into the underlying quantum theory, but says the behavior between the two regimes is only the most natural possibility.
The most natural possibility is that it stays abrupt everywhere in between.
Source-stated
Anthropic reports that Claude found and proved the optimal ceiling for the simplest version of the modular bootstrap, using a spectral curve.
Claude found the spectral curve that governs the modular bootstrap at large values of c. It then used it to prove that the lightest black hole in 3D quantum gravity can be no heavier than c/Λ, where Λ is given by an exact formula and is approximately 9.1266. It also proved that this ceiling is opti
Source-stated
Anthropic says this result does not identify the correct theory and that lowering the ceiling to the mass at which black holes first appear will require other ingredients.
It also shows that this version of the method can’t, on its own, push the ceiling down to c/12, the mass at which black holes first appear; that will take other ingredients.
Source-stated
Anthropic reports that agents worked in the background for about six weeks; the modular bootstrap result was formally verified in Lean and a physicist checked every main result.
The agents ran in the background, with no special setup, for about six weeks.
Source-stated
Anthropic says neither result describes our universe directly or has practical applications.
Neither result describes our own universe directly, and neither has practical applications
Source and contributor context
Anthropic · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
Media declared by the original publisher; availability checked separately from the editorial reading.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Formal verification
Using software to check every step of a proof or a program against precise rules.
Anthropic
Anthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Claude
Claude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Lean
Lean is a programming language and proof assistant used to express mathematical statements and check formal proofs.
Editorial provenance
Version 1. Publication ID: 2859af70-3dff-4319-843b-e14bfdd8b4bb.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-18T20:02:40.772Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Selected after a second opinion by claude-opus-5-5, because the admission classifier's confidence was under its bar: Reports AI agents autonomously solving two open physics problems over ~6 weeks, cross-checked by a physicist's independent result and Lean verification, after prior failed attempts: a capability step.
Observation: 412763ebf1d7e8d8268cd426dad2e609e004dd524115a366f8a4537b276bae37. Source text hash: 539ee2df40e03cf0d0c3fa061463b0ac0fc1c6053e4401a689204419941c2e4a.
Revision and source-check history
Version 1 · 2026-11-18T20:02:40.772Z
Anthropic says Claude solved two open problems in black hole physics
reports that completed a long-standing calculation about whether a phase transition remains abrupt in a quantum theory, and established an exact limit for a technique used to rule out candidate theories of quantum gravity. The results concern theoretical model universes, not our own universe directly, and Anthropic says neither has practical applications. For the first problem, Anthropic says the transition is abrupt both in the regime described by Hawking and Page and in a regime where spacetime loses a definite shape. It says this suggests black-hole formation is built into the underlying quantum theory, while the behavior between those regimes remains unestablished. For the second, Claude used a spectral curve to determine the optimal ceiling for the simplest version of the modular bootstrap. The result narrows the search but does not identify the correct theory. Anthropic says agents worked on the problems for about six weeks, and that the modular bootstrap result was in and checked by a physicist. The two results involve different work: completing a difficult calculation and using a mathematical tool to establish a limit.
Original source
· Article
Source details
Original source date: Nov 17, 2026 (2026-11-17T00:00:00.000Z).
Date basis: Date stated by the publisher.
Added to Escape Meridian: 2026-11-18T20:02:40.772Z. First observed: 2026-11-18T18:04:05.367Z. Retrieved: 2026-11-18T19:47:01.288Z.
Limits
describes theoretical model universes; it says neither result describes our own universe directly or has practical applications. The first result establishes the abrupt transition at the two studied extremes; Anthropic presents persistence between them as the most natural possibility, not as established. The modular bootstrap result concerns the simplest version of the method, which ignores how black holes spin, and does not identify the correct theory.
Open question
What further ingredients would be needed to push the modular bootstrap ceiling lower?
Source-stated
Anthropic reports that Claude completed a calculation finding the phase transition abrupt in a quantum theory, as well as in the regime described by Hawking and Page.
It showed that the phase transition is abrupt. The same sudden switch appears at both extremes
Source-stated
Anthropic says the result suggests black-hole formation is built into the underlying quantum theory, but says the behavior between the two regimes is only the most natural possibility.
The most natural possibility is that it stays abrupt everywhere in between.
Source-stated
Anthropic reports that Claude found and proved the optimal ceiling for the simplest version of the modular bootstrap, using a spectral curve.
Claude found the spectral curve that governs the modular bootstrap at large values of c. It then used it to prove that the lightest black hole in 3D quantum gravity can be no heavier than c/Λ, where Λ is given by an exact formula and is approximately 9.1266. It also proved that this ceiling is opti
Source-stated
Anthropic says this result does not identify the correct theory and that lowering the ceiling to the mass at which black holes first appear will require other ingredients.
It also shows that this version of the method can’t, on its own, push the ceiling down to c/12, the mass at which black holes first appear; that will take other ingredients.
Source-stated
Anthropic reports that agents worked in the background for about six weeks; the modular bootstrap result was formally verified in Lean and a physicist checked every main result.
The agents ran in the background, with no special setup, for about six weeks.
Source-stated
Anthropic says neither result describes our universe directly or has practical applications.
Neither result describes our own universe directly, and neither has practical applications
Source and contributor context
Anthropic · Article author
These are credits stated by the source; they do not establish a profile match or authorship of every passage.
Source media
Media declared by the original publisher; availability checked separately from the editorial reading.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Formal verification
Using software to check every step of a proof or a program against precise rules.
Anthropic
Anthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Claude
Claude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Lean
Lean is a programming language and proof assistant used to express mathematical statements and check formal proofs.
Editorial provenance
Version 1. Publication ID: 2859af70-3dff-4319-843b-e14bfdd8b4bb.
Prepared by gpt-6-luna; grounding review by gpt-6-luna.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-18T20:02:40.772Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: material_development.
Selected after a second opinion by claude-opus-5-5, because the admission classifier's confidence was under its bar: Reports AI agents autonomously solving two open physics problems over ~6 weeks, cross-checked by a physicist's independent result and Lean verification, after prior failed attempts: a capability step.
Observation: 412763ebf1d7e8d8268cd426dad2e609e004dd524115a366f8a4537b276bae37. Source text hash: 539ee2df40e03cf0d0c3fa061463b0ac0fc1c6053e4401a689204419941c2e4a.
uncertain · 2026-11-18T20:02:40.772Z · anthropic/claude-opus-5.5
Publication: 2859af70-3dff-4319-843b-e14bfdd8b4bb
attribution — pass: The reading consistently frames the results as Anthropic's report about Claude, using phrases like 'Anthropic reports' and 'Anthropic says'. It does not present the physics claims as established fact or as Escape Meridian's view. The publisher is correctly identified.
Claude has solved two open problems in theoretical physics.
summary — uncertain: The summary is mostly faithful, but two gaps affect meaning. First, 'the regime described by Hawking and Page' is not explained; a lay reader will not know it is the large, smooth-spacetime end. Repair: add a few words on that. Second, it drops what the source stresses. Claude spotted a pattern physicists had missed, and the work was 'largely autonomously' done. Half of the final proof follows an OpenAI model's proof strategy. An independent physicist reached the same Hawking-Page answer by a different method. Repair: one clause on autonomy and one crediting the OpenAI-derived proof strategy. 'Lean' also needs a gloss as a computerized proof checker.
One half of the final proof, showing that the ceiling can’t be lowered, follows the strategy of the OpenAI proof.
date — pass: The value in evidence.originalPublished, 2026-11-17, comes from the publisher page. It matches the feed date and precedes the first observation. Nothing indicates an earlier first publication.
support — pass: Each claim matches the source text. The source supports: an abrupt transition at both extremes; 'most natural possibility' in between; the exact ceiling of about 9.1266 proved optimal; c/12 requiring other ingredients; about six weeks; Lean verification and the physicist check; and no practical applications. The limitations accurately preserve scope, including that the result ignores spin. The question follows directly from the source's statement that other ingredients will be needed. No unsupported benefits or costs are implied.
In September, the results were formally verified in Lean, a computerized proof checker. Claude then wrote the paper with minimal guidance, and a physicist checked every main result.
media — pass: The selected media is the verified direct link to the original Anthropic research post. This is the most direct way to reach the source.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
Anthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore AnthropicAnthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore AnthropicClaude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Explore ClaudeUsing software to check every step of a proof or a program against precise rules.
Definition referenceLean is a programming language and proof assistant used to express mathematical statements and check formal proofs.
Explore LeanAnthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore AnthropicAnthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore AnthropicAnthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore AnthropicClaude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.
Explore ClaudeUsing software to check every step of a proof or a program against precise rules.
Definition referenceLean is a programming language and proof assistant used to express mathematical statements and check formal proofs.
Explore LeanAnthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore AnthropicAnthropic develops Claude and publishes research about AI capability, safety and how people use its systems.
Explore Anthropic