{"id":"6ab635fc0b01f5e4dd5703b1d35acd7dbd5fba0675486ddd816b734ee3e6d848","title":"Matt von Hippel argues in Anthropic that Claude’s calculation changed where he sees the limit","argument":"Originally published September 25, 2026. Matt von Hippel, writing for Anthropic, describes how Anthropic physicists used Claude to calculate a nine-loop amplitude in the toy model N=4 super Yang-Mills. The result was independently checked by Lance Dixon, and von Hippel says Claude used known methods with more computing power than people had tried before—not a super-intelligent new approach. His argument is that the apparent computational barrier was lower than he had assumed: the calculation was within reach on a reasonable budget, and Claude carried it through without outside scientific input. Dixon adds that the calculation used methods developed by his collaborators, and says Claude had to build the code for a fragile, complicated recipe. Von Hippel’s takeaway is that experts may overlook achievable progress, and that scientists should consider whether AI can tackle other frontier calculations while planning how to check results. But he says this case does not answer his broader question about AI’s future or superintelligence; he had hoped to see new methods and a glimpse of how far AI could push computational limits. Dixon, meanwhile, says the more searching moment will come if large language models begin producing new physical principles and insights before humans.","topicIds":["artificial-general-intelligence"],"domain":"other","claimType":"attributed reading","limits":"Text as retrieved on October 30, 2026; originally published September 25, 2026. Anthropic invited Matt von Hippel to write the post and compensated him; Anthropic staff gave feedback on drafts. The disclosure says the content and opinions are his own. Lance Dixon independently validated the result and received Claude usage credits. Von Hippel says he is unsure how far the result generalizes beyond this toy model calculation.","speaker":{"name":"Anthropic","roleAtSource":"Publisher"},"source":{"title":"Claude computes a nine-loop amplitude in N=4 super-Yang-Mills","publisher":"Anthropic","format":"Article","url":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","publishedAt":"2026-09-25T00:00:00.000Z","dateBasis":"Date stated on the publisher’s page."},"evidence":[{"text":"Matt von Hippel says Anthropic’s Claude computed a nine-loop amplitude using known methods and more computing power than people had previously tried.","excerpt":"Claude used known methods, with a bit more compute than people had tried to use before.","url":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","relationship":"Source-stated"},{"text":"Von Hippel says the result changed his view of how achievable the calculation was, but did not give him the broader insight about AI’s future that he had hoped for.","excerpt":"I wanted to know how far AI could push computational limits… and I feel like what I learned here is just that I was too naïve about where the limit was.","url":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","relationship":"Source-stated"},{"text":"Lance Dixon says Claude used methods his collaborators developed, and that the calculation required Claude to build the code for a fragile recipe.","excerpt":"So Claude had to develop all that code from scratch.","url":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","relationship":"Source-stated"},{"text":"Dixon says a more searching moment will come if large language models produce new physical principles and insights before humans.","excerpt":"But the more soul-searching moments will come when large language models start to come up with new physical principles and insights before humans.","url":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","relationship":"Source-stated"}],"sourceMedia":{"status":"unchecked","sourceUrl":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","checkedAt":null,"items":[]},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Can AI use similar approaches to advance real-world amplitudes calculations, and how should researchers check the results?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"Matt von Hippel argues in Anthropic that Claude’s calculation changed where he sees the limit","summary":"Originally published September 25, 2026. Matt von Hippel, writing for Anthropic, describes how Anthropic physicists used Claude to calculate a nine-loop amplitude in the toy model N=4 super Yang-Mills. The result was independently checked by Lance Dixon, and von Hippel says Claude used known methods with more computing power than people had tried before—not a super-intelligent new approach. His argument is that the apparent computational barrier was lower than he had assumed: the calculation was within reach on a reasonable budget, and Claude carried it through without outside scientific input. Dixon adds that the calculation used methods developed by his collaborators, and says Claude had to build the code for a fragile, complicated recipe. Von Hippel’s takeaway is that experts may overlook achievable progress, and that scientists should consider whether AI can tackle other frontier calculations while planning how to check results. But he says this case does not answer his broader question about AI’s future or superintelligence; he had hoped to see new methods and a glimpse of how far AI could push computational limits. Dixon, meanwhile, says the more searching moment will come if large language models begin producing new physical principles and insights before humans.","question":"Can AI use similar approaches to advance real-world amplitudes calculations, and how should researchers check the results?","claims":[{"text":"Matt von Hippel says Anthropic’s Claude computed a nine-loop amplitude using known methods and more computing power than people had previously tried.","excerpt":"Claude used known methods, with a bit more compute than people had tried to use before.","support":"source-stated"},{"text":"Von Hippel says the result changed his view of how achievable the calculation was, but did not give him the broader insight about AI’s future that he had hoped for.","excerpt":"I wanted to know how far AI could push computational limits… and I feel like what I learned here is just that I was too naïve about where the limit was.","support":"source-stated"},{"text":"Lance Dixon says Claude used methods his collaborators developed, and that the calculation required Claude to build the code for a fragile recipe.","excerpt":"So Claude had to develop all that code from scratch.","support":"source-stated"},{"text":"Dixon says a more searching moment will come if large language models produce new physical principles and insights before humans.","excerpt":"But the more soul-searching moments will come when large language models start to come up with new physical principles and insights before humans.","support":"source-stated"}],"limitations":["Text as retrieved on October 30, 2026; originally published September 25, 2026.","Anthropic invited Matt von Hippel to write the post and compensated him; Anthropic staff gave feedback on drafts. The disclosure says the content and opinions are his own. Lance Dixon independently validated the result and received Claude usage credits.","Von Hippel says he is unsure how far the result generalizes beyond this toy model calculation."],"mediaId":null,"correctionReason":null,"id":"6ab635fc0b01f5e4dd5703b1d35acd7dbd5fba0675486ddd816b734ee3e6d848","publicationId":"ea3e1e25-82d9-70d9-45dd-eef2a61576c9","version":1,"sourceUrl":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","sourceTitle":"Claude computes a nine-loop amplitude in N=4 super-Yang-Mills","publisher":"Anthropic","publishedAt":"2026-09-25T00:00:00.000Z","originalPublished":{"value":"2026-09-25T00:00:00.000Z","precision":"instant","basis":"backfill:article_published"},"firstObservedAt":"2026-10-18T07:29:19.309Z","createdAt":"2026-11-05T14:13:23.621Z","historical":true,"entityIds":["anthropic"],"topicIds":["artificial-general-intelligence"],"media":[],"sourceMedia":{"status":"unchecked","sourceUrl":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","checkedAt":null,"items":[]},"editorialDecision":{"outcome":"accept","mediaId":null,"inputHash":"ed53b096ccf3412489a0f80eb5994afbc7fe695fa196c22714daa78ac79f481c","rubricVersion":"unrecorded","provider":"typesafe","model":"jev-1.13.0","reasonCodes":["bridge_admitted_by_model"],"confidence":0.61,"decidedAt":"2026-10-30T03:37:50.848Z"},"provenance":{"textHash":"0b14667501a883103183280363b566b9731252b78a7caaad76266deb03c724e3","observationId":"b47e38559a585cce9fb3cfb8050a3ba31d91656c51aff69a620443b0f7f5346a","retrievedAt":"2026-10-30T03:55:44.816Z","kind":"publisher-page","preparer":{"model":"openai/gpt-6-luna","runId":"195bd768-4519-aa6d-aec6-fd6eb89dac8d"},"reviewer":{"model":"anthropic/claude-opus-5.5","runId":"b9f6873f-082d-36cc-fa41-d1f7b242048e"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-05T14:08:46.499Z","model":"anthropic/claude-opus-5.5"}},"canonicalUrl":"https://escapemeridian.com/readings/matt-von-hippel-argues-in-anthropic-that-claude-s-calculation-6ab635fc","formats":{"html":"/readings/matt-von-hippel-argues-in-anthropic-that-claude-s-calculation-6ab635fc","markdown":"/readings/matt-von-hippel-argues-in-anthropic-that-claude-s-calculation-6ab635fc.md","json":"/readings/matt-von-hippel-argues-in-anthropic-that-claude-s-calculation-6ab635fc.json"},"attribution":"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.","maintainedGuideLinks":[],"readerContext":{"version":"reader-context-v1","attribution":"Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.","terms":[{"id":"term-superintelligence","name":"Superintelligence","explanation":"A hypothetical AI far more capable than people across most intellectual tasks.","referenceUrl":"https://en.wikipedia.org/wiki/Superintelligence"},{"id":"term-llm","name":"Large language model","explanation":"A language model with many learned numerical settings, used to process and generate language.","referenceUrl":"https://developers.google.com/machine-learning/glossary#large-language-model"}],"entities":[{"id":"anthropic","name":"Anthropic","kind":"organization","context":"Anthropic develops Claude and publishes research about AI capability, safety and how people use its systems.","referenceUrls":["https://www.anthropic.com/news","https://www.anthropic.com/research/team/economics"],"path":"/entities/anthropic","json":"/entities/anthropic.json","markdown":"/entities/anthropic.md"},{"id":"claude","name":"Claude","kind":"model","context":"Claude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.","referenceUrls":["https://claude.com/product/overview","https://www.anthropic.com/news","https://www.anthropic.com/research/team/economics"],"path":"/entities/claude","json":"/entities/claude.json","markdown":"/entities/claude.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"6ab635fc0b01f5e4dd5703b1d35acd7dbd5fba0675486ddd816b734ee3e6d848","revision":"ea3e1e25-82d9-70d9-45dd-eef2a61576c9","revisionKind":"publication-id","title":"Matt von Hippel argues in Anthropic that Claude’s calculation changed where he sees the limit","url":"https://escapemeridian.com/readings/matt-von-hippel-argues-in-anthropic-that-claude-s-calculation-6ab635fc"},"source":{"type":"source","id":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","revision":"0b14667501a883103183280363b566b9731252b78a7caaad76266deb03c724e3","revisionKind":"sha256-source-text","title":"Claude computes a nine-loop amplitude in N=4 super-Yang-Mills","url":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter."},"revisions":[{"id":"6ab635fc0b01f5e4dd5703b1d35acd7dbd5fba0675486ddd816b734ee3e6d848","title":"Matt von Hippel argues in Anthropic that Claude’s calculation changed where he sees the limit","argument":"Originally published September 25, 2026. Matt von Hippel, writing for Anthropic, describes how Anthropic physicists used Claude to calculate a nine-loop amplitude in the toy model N=4 super Yang-Mills. The result was independently checked by Lance Dixon, and von Hippel says Claude used known methods with more computing power than people had tried before—not a super-intelligent new approach. His argument is that the apparent computational barrier was lower than he had assumed: the calculation was within reach on a reasonable budget, and Claude carried it through without outside scientific input. Dixon adds that the calculation used methods developed by his collaborators, and says Claude had to build the code for a fragile, complicated recipe. Von Hippel’s takeaway is that experts may overlook achievable progress, and that scientists should consider whether AI can tackle other frontier calculations while planning how to check results. But he says this case does not answer his broader question about AI’s future or superintelligence; he had hoped to see new methods and a glimpse of how far AI could push computational limits. Dixon, meanwhile, says the more searching moment will come if large language models begin producing new physical principles and insights before humans.","topicIds":["artificial-general-intelligence"],"domain":"other","claimType":"attributed reading","limits":"Text as retrieved on October 30, 2026; originally published September 25, 2026. Anthropic invited Matt von Hippel to write the post and compensated him; Anthropic staff gave feedback on drafts. The disclosure says the content and opinions are his own. Lance Dixon independently validated the result and received Claude usage credits. Von Hippel says he is unsure how far the result generalizes beyond this toy model calculation.","speaker":{"name":"Anthropic","roleAtSource":"Publisher"},"source":{"title":"Claude computes a nine-loop amplitude in N=4 super-Yang-Mills","publisher":"Anthropic","format":"Article","url":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","publishedAt":"2026-09-25T00:00:00.000Z","dateBasis":"Date stated on the publisher’s page."},"evidence":[{"text":"Matt von Hippel says Anthropic’s Claude computed a nine-loop amplitude using known methods and more computing power than people had previously tried.","excerpt":"Claude used known methods, with a bit more compute than people had tried to use before.","url":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","relationship":"Source-stated"},{"text":"Von Hippel says the result changed his view of how achievable the calculation was, but did not give him the broader insight about AI’s future that he had hoped for.","excerpt":"I wanted to know how far AI could push computational limits… and I feel like what I learned here is just that I was too naïve about where the limit was.","url":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","relationship":"Source-stated"},{"text":"Lance Dixon says Claude used methods his collaborators developed, and that the calculation required Claude to build the code for a fragile recipe.","excerpt":"So Claude had to develop all that code from scratch.","url":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","relationship":"Source-stated"},{"text":"Dixon says a more searching moment will come if large language models produce new physical principles and insights before humans.","excerpt":"But the more soul-searching moments will come when large language models start to come up with new physical principles and insights before humans.","url":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","relationship":"Source-stated"}],"sourceMedia":{"status":"unchecked","sourceUrl":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","checkedAt":null,"items":[]},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Can AI use similar approaches to advance real-world amplitudes calculations, and how should researchers check the results?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"Matt von Hippel argues in Anthropic that Claude’s calculation changed where he sees the limit","summary":"Originally published September 25, 2026. Matt von Hippel, writing for Anthropic, describes how Anthropic physicists used Claude to calculate a nine-loop amplitude in the toy model N=4 super Yang-Mills. The result was independently checked by Lance Dixon, and von Hippel says Claude used known methods with more computing power than people had tried before—not a super-intelligent new approach. His argument is that the apparent computational barrier was lower than he had assumed: the calculation was within reach on a reasonable budget, and Claude carried it through without outside scientific input. Dixon adds that the calculation used methods developed by his collaborators, and says Claude had to build the code for a fragile, complicated recipe. Von Hippel’s takeaway is that experts may overlook achievable progress, and that scientists should consider whether AI can tackle other frontier calculations while planning how to check results. But he says this case does not answer his broader question about AI’s future or superintelligence; he had hoped to see new methods and a glimpse of how far AI could push computational limits. Dixon, meanwhile, says the more searching moment will come if large language models begin producing new physical principles and insights before humans.","question":"Can AI use similar approaches to advance real-world amplitudes calculations, and how should researchers check the results?","claims":[{"text":"Matt von Hippel says Anthropic’s Claude computed a nine-loop amplitude using known methods and more computing power than people had previously tried.","excerpt":"Claude used known methods, with a bit more compute than people had tried to use before.","support":"source-stated"},{"text":"Von Hippel says the result changed his view of how achievable the calculation was, but did not give him the broader insight about AI’s future that he had hoped for.","excerpt":"I wanted to know how far AI could push computational limits… and I feel like what I learned here is just that I was too naïve about where the limit was.","support":"source-stated"},{"text":"Lance Dixon says Claude used methods his collaborators developed, and that the calculation required Claude to build the code for a fragile recipe.","excerpt":"So Claude had to develop all that code from scratch.","support":"source-stated"},{"text":"Dixon says a more searching moment will come if large language models produce new physical principles and insights before humans.","excerpt":"But the more soul-searching moments will come when large language models start to come up with new physical principles and insights before humans.","support":"source-stated"}],"limitations":["Text as retrieved on October 30, 2026; originally published September 25, 2026.","Anthropic invited Matt von Hippel to write the post and compensated him; Anthropic staff gave feedback on drafts. The disclosure says the content and opinions are his own. Lance Dixon independently validated the result and received Claude usage credits.","Von Hippel says he is unsure how far the result generalizes beyond this toy model calculation."],"mediaId":null,"correctionReason":null,"id":"6ab635fc0b01f5e4dd5703b1d35acd7dbd5fba0675486ddd816b734ee3e6d848","publicationId":"ea3e1e25-82d9-70d9-45dd-eef2a61576c9","version":1,"sourceUrl":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","sourceTitle":"Claude computes a nine-loop amplitude in N=4 super-Yang-Mills","publisher":"Anthropic","publishedAt":"2026-09-25T00:00:00.000Z","originalPublished":{"value":"2026-09-25T00:00:00.000Z","precision":"instant","basis":"backfill:article_published"},"firstObservedAt":"2026-10-18T07:29:19.309Z","createdAt":"2026-11-05T14:13:23.621Z","historical":true,"entityIds":["anthropic"],"topicIds":["artificial-general-intelligence"],"media":[],"sourceMedia":{"status":"unchecked","sourceUrl":"https://www.anthropic.com/research/yes-claude-can-do-nine-loops","checkedAt":null,"items":[]},"editorialDecision":{"outcome":"accept","mediaId":null,"inputHash":"ed53b096ccf3412489a0f80eb5994afbc7fe695fa196c22714daa78ac79f481c","rubricVersion":"unrecorded","provider":"typesafe","model":"jev-1.13.0","reasonCodes":["bridge_admitted_by_model"],"confidence":0.61,"decidedAt":"2026-10-30T03:37:50.848Z"},"provenance":{"textHash":"0b14667501a883103183280363b566b9731252b78a7caaad76266deb03c724e3","observationId":"b47e38559a585cce9fb3cfb8050a3ba31d91656c51aff69a620443b0f7f5346a","retrievedAt":"2026-10-30T03:55:44.816Z","kind":"publisher-page","preparer":{"model":"openai/gpt-6-luna","runId":"195bd768-4519-aa6d-aec6-fd6eb89dac8d"},"reviewer":{"model":"anthropic/claude-opus-5.5","runId":"b9f6873f-082d-36cc-fa41-d1f7b242048e"}},"audit":{"status":"uncertain","reviewedAt":"2026-11-05T14:08:46.499Z","model":"anthropic/claude-opus-5.5"}}}],"audits":[{"publicationId":"ea3e1e25-82d9-70d9-45dd-eef2a61576c9","verdict":"uncertain","findings":[{"dimension":"attribution","status":"pass","explanation":"The reading correctly attributes the post to Matt von Hippel, writing for Anthropic at Anthropic's invitation. It attributes the addendum views to Lance Dixon. It keeps the claims as the authors' assertions and notes the compensation disclosure.","excerpt":"Anthropic invited Matt von Hippel to write this post and compensated him for his time."},{"dimension":"summary","status":"uncertain","explanation":"The central terms 'nine-loop amplitude', 'loop' and 'N=4 super Yang-Mills' are never explained, and the definitions cover only superintelligence and LLM. A general reader cannot tell what was computed or why nine loops is hard. Repair: add a plain gloss, e.g. scattering amplitudes predict particle reactions and loops measure their complexity. The summary also omits that Song He's group nearly reached the result, which grounds von Hippel's point that humans could do it too.","excerpt":"Instead, it did something it turned out humans were also able to do."},{"dimension":"date","status":"pass","explanation":"The reading's September 25, 2026 date matches the UTC day of originalPublished. The source's timeline (end of August contact, a September 1 notification, two weeks of validation) is consistent with it. There is no hindsight framing, and the kind is publisher-page, so no archive disclosure is needed.","excerpt":"For me, it happened on September 1, when Liam Fitzpatrick and Siddharth Mishra-Sharma at Anthropic told me"},{"dimension":"support","status":"uncertain","explanation":"Most claims match the source. However, claim 4 and the summary turn Dixon's confident 'when' into a conditional 'if', which weakens his forecast that LLMs will produce new physical principles. Smallest repair: write 'will come when'. Also, the page was edited after publication (dateModified 2026-10-21), so it is not plainly confirmed that the cited passages, including the disclosure, are original.","excerpt":"But the more soul-searching moments will come when large language models start to come up with new physical principles and insights before humans."},{"dimension":"media","status":"pass","explanation":"No media was selected, and no verified media candidates were supplied, so leaving media empty is appropriate.","excerpt":null}],"createdAt":"2026-11-05T14:08:46.499Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}