{"schema_version":"onlylabs.public_signal.v1","title":"Anthropic Writing: Formal Verification","description":"Anthropic writing signal with public source context, captured evidence pages, related signals, and data-business radar classification.","url":"https://onlylabs.fyi/signals/59425ae9-c241-48af-84ca-ae973091b1c1","json_url":"https://onlylabs.fyi/signals/59425ae9-c241-48af-84ca-ae973091b1c1/signal.json","generated_at":"2026-10-25T03:51:16.987Z","evidence_latest_fetched_at":"2026-10-23T00:03:18.663+00:00","signal_first_seen_at":"2026-10-23T00:00:49.326198+00:00","org":{"slug":"anthropic","name":"Anthropic","category":"frontier-lab","category_label":"Frontier lab","dossier_url":"https://onlylabs.fyi/labs/anthropic","dossier_json_url":"https://onlylabs.fyi/labs/anthropic/dossier.json"},"related_urls":{"signal":"https://onlylabs.fyi/signals/59425ae9-c241-48af-84ca-ae973091b1c1","signal_json":"https://onlylabs.fyi/signals/59425ae9-c241-48af-84ca-ae973091b1c1/signal.json","source":"https://www.anthropic.com/research/formal-verification","lab_dossier":"https://onlylabs.fyi/labs/anthropic","lab_dossier_json":"https://onlylabs.fyi/labs/anthropic/dossier.json","analysis":"https://onlylabs.fyi/analysis/anthropic","analysis_json":"https://onlylabs.fyi/analysis/anthropic/analysis.json","analysis_evidence_json":"https://onlylabs.fyi/analysis/anthropic/evidence.json","category":"https://onlylabs.fyi/frontier","category_json":"https://onlylabs.fyi/frontier.json","category_feed":"https://onlylabs.fyi/frontier/feed.xml","category_signals_json":"https://onlylabs.fyi/signals.json","topic":"https://onlylabs.fyi/topics/talking","topic_signals_json":"https://onlylabs.fyi/topics/talking/signals.json","topic_feed":"https://onlylabs.fyi/topics/talking/feed.xml","data_business":null},"answer_pack":{"answer":"Anthropic published Formal Verification. This talking signal gives public context for research themes, product direction, policy, or launch framing. High-signal details: Anthropic formal verification post, low HN engagement. · Comprehensive Formal Verification of an OS Microkernel GERWIN KLEIN, JUNE ANDRONICK, KEVIN ELPHINSTONE, TOBY MURRAY, THOMAS SEWELL, RAFAL KOLANSKI, and GERNOT HEISER,.... onlylabs links this event to 4 captured evidence pages and 6 related writing signals.","signal_desk":"talking","source_context":{"source_url":"https://www.anthropic.com/research/formal-verification","source_host":"anthropic.com","occurred_at":"2026-10-23T20:56:14+00:00","first_seen_at":"2026-10-23T00:00:49.326198+00:00","date_source":"sitemap.lastmod","context":null},"context_markers":[{"label":"Lab","value":"Anthropic","source":"signal"},{"label":"Signal desk","value":"talking","source":"signal"},{"label":"Source host","value":"anthropic.com","source":"source"},{"label":"PDF","value":"linked report","source":"source"},{"label":"Notability","value":"Anthropic formal verification post, low HN engagement.","source":"signal"},{"label":"HN","value":"Skepticism that LLMs can scale formal verification proofs, supporting the convex hull hypothesis.","source":"source"},{"label":"Watch term","value":"Infrastructure","source":"evidence"},{"label":"Watch term","value":"Safety and alignment","source":"evidence"},{"label":"Watch term","value":"Agents and tool use","source":"evidence"}],"evidence_coverage":{"target_pages":4,"captured_pages":4,"readable_pages":4,"capture_methods":["exa","plain"],"missing_page_urls":[],"failed_page_urls":[],"blocked_page_urls":[],"page_urls":["https://www.anthropic.com/research/formal-verification","https://sel4.systems/Research/pdfs/comprehensive-formal-verification-os-microkernel.pdf","https://www.astree.ens.fr/papers/astree_airbus_sas2007.pdf","https://www-cdn.anthropic.com/files/4zrzovbb/website/6867eadf760258fa8b42a504600cd3c036d16358.pdf"],"related_signals":6,"has_source_url":true,"latest_page_fetched_at":"2026-10-23T00:03:18.663+00:00"},"data_business":{"matches":false,"lanes":[],"matched_terms":[],"score":null,"reason":null},"agent_handoff":{"signal_json":"https://onlylabs.fyi/signals/59425ae9-c241-48af-84ca-ae973091b1c1/signal.json","dossier_json":"https://onlylabs.fyi/labs/anthropic/dossier.json","analysis_json":"https://onlylabs.fyi/analysis/anthropic/analysis.json","analysis_evidence_json":"https://onlylabs.fyi/analysis/anthropic/evidence.json","topic_signals_json":"https://onlylabs.fyi/topics/talking/signals.json","topic_feed":"https://onlylabs.fyi/topics/talking/feed.xml","category_signals_json":"https://onlylabs.fyi/signals.json","data_radar_json":null,"opportunities_json":null},"analysis_playbook":{"objective":"Turn public writing and discussion into a readable map of research themes, product framing, policy posture, launch narratives, and market attention.","evidence_focus":["post title","source URL","captured page text","HN traction","linked model or paper references","publication date"],"extraction_questions":["Which themes are labs choosing to explain publicly?","Which posts are attracting outside discussion?","Which writing reframes a recent release, model, hiring wave, or policy stance?","Which posts mention data, evals, infrastructure, safety, or deployment workflows?"],"signal_questions":["What public theme, launch framing, or research direction does this writing signal expose?","Which themes are labs choosing to explain publicly?","Which posts are attracting outside discussion?","Do the 6 related writing signals show a repeated pattern?"],"output_fields":["org","theme","public_framing","traction","data_business_lane","evidence_url"],"data_business_relevance":"Public writing supplies the narrative layer over raw signals and helps identify which frontier-lab priorities are becoming externally legible.","required_sources":[{"label":"signal_json","url":"https://onlylabs.fyi/signals/59425ae9-c241-48af-84ca-ae973091b1c1/signal.json","required":true},{"label":"source","url":"https://www.anthropic.com/research/formal-verification","required":true},{"label":"dossier_json","url":"https://onlylabs.fyi/labs/anthropic/dossier.json","required":true},{"label":"analysis_evidence_json","url":"https://onlylabs.fyi/analysis/anthropic/evidence.json","required":true},{"label":"topic_signals_json","url":"https://onlylabs.fyi/topics/talking/signals.json","required":false},{"label":"data_radar_json","url":null,"required":false}],"expected_output":["one-paragraph source-grounded interpretation","category-specific implication","confidence and missing evidence","recommended next source to inspect"],"prompt_seed":"Using only the linked onlylabs JSON, captured source context, and cited evidence, analyze Anthropic's writing signal \"Formal Verification\" for frontier lab strategy."},"semantic_triples":[{"subject":"Anthropic","predicate":"published","object":"Formal Verification","text":"Anthropic published Formal Verification."},{"subject":"Formal Verification","predicate":"is classified as","object":"writing signal","text":"Formal Verification is classified as writing signal."},{"subject":"Formal Verification","predicate":"belongs to","object":"talking desk","text":"Formal Verification belongs to talking desk."},{"subject":"Formal Verification","predicate":"has evidence coverage","object":"4 captured evidence pages","text":"Formal Verification has evidence coverage 4 captured evidence pages."},{"subject":"Formal Verification","predicate":"has captured page count","object":"4","text":"Formal Verification has captured page count 4."},{"subject":"Formal Verification","predicate":"has readable page count","object":"4","text":"Formal Verification has readable page count 4."},{"subject":"Formal Verification","predicate":"has related signal count","object":"6","text":"Formal Verification has related signal count 6."},{"subject":"Formal Verification","predicate":"has analysis playbook objective","object":"Turn public writing and discussion into a readable map of research themes, product framing, policy posture, launch narratives, and market attention.","text":"Formal Verification has analysis playbook objective Turn public writing and discussion into a readable map of research themes, product framing, policy posture, launch narratives, and market attention.."},{"subject":"Formal Verification","predicate":"has source host","object":"anthropic.com","text":"Formal Verification has source host anthropic.com."},{"subject":"Formal Verification","predicate":"has lab","object":"Anthropic","text":"Formal Verification has lab Anthropic."},{"subject":"Formal Verification","predicate":"has signal desk","object":"talking","text":"Formal Verification has signal desk talking."},{"subject":"Formal Verification","predicate":"has source host","object":"anthropic.com","text":"Formal Verification has source host anthropic.com."},{"subject":"Formal Verification","predicate":"has pdf","object":"linked report","text":"Formal Verification has pdf linked report."},{"subject":"Formal Verification","predicate":"has notability","object":"Anthropic formal verification post, low HN engagement.","text":"Formal Verification has notability Anthropic formal verification post, low HN engagement.."},{"subject":"Formal Verification","predicate":"has hn","object":"Skepticism that LLMs can scale formal verification proofs, supporting the convex hull hypothesis.","text":"Formal Verification has hn Skepticism that LLMs can scale formal verification proofs, supporting the convex hull hypothesis.."},{"subject":"Formal Verification","predicate":"has watch term","object":"Infrastructure","text":"Formal Verification has watch term Infrastructure."},{"subject":"Formal Verification","predicate":"has watch term","object":"Safety and alignment","text":"Formal Verification has watch term Safety and alignment."}]},"intelligence":{"signal_desk":"talking","answer":"Anthropic published Formal Verification. This talking signal gives public context for research themes, product direction, policy, or launch framing. High-signal details: Anthropic formal verification post, low HN engagement. · Comprehensive Formal Verification of an OS Microkernel GERWIN KLEIN, JUNE ANDRONICK, KEVIN ELPHINSTONE, TOBY MURRAY, THOMAS SEWELL, RAFAL KOLANSKI, and GERNOT HEISER,.... onlylabs links this event to 4 captured evidence pages and 6 related writing signals.","semantic_triples":[{"subject":"Anthropic","predicate":"published","object":"Formal Verification","text":"Anthropic published Formal Verification."},{"subject":"Formal Verification","predicate":"is classified as","object":"writing signal","text":"Formal Verification is classified as writing signal."},{"subject":"Formal Verification","predicate":"belongs to","object":"talking desk","text":"Formal Verification belongs to talking desk."},{"subject":"Formal Verification","predicate":"has evidence coverage","object":"4 captured evidence pages","text":"Formal Verification has evidence coverage 4 captured evidence pages."}]},"signal":{"id":"59425ae9-c241-48af-84ca-ae973091b1c1","url":"https://onlylabs.fyi/signals/59425ae9-c241-48af-84ca-ae973091b1c1","json_url":"https://onlylabs.fyi/signals/59425ae9-c241-48af-84ca-ae973091b1c1/signal.json","source_url":"https://www.anthropic.com/research/formal-verification","title":"Formal Verification","summary":"Anthropic published a writing signal. onlylabs watches public writing for research themes, product direction, and model-launch context.","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"anthropic","name":"Anthropic","category":"frontier-lab"},"occurred_at":"2026-10-23T20:56:14+00:00","first_seen_at":"2026-10-23T00:00:49.326198+00:00","date_source":"sitemap.lastmod","evidence_coverage":{"target_pages":4,"captured_pages":4,"readable_pages":4,"capture_methods":["exa","plain"],"missing_page_urls":[],"failed_page_urls":[],"blocked_page_urls":[],"page_urls":["https://www.anthropic.com/research/formal-verification","https://sel4.systems/Research/pdfs/comprehensive-formal-verification-os-microkernel.pdf","https://www.astree.ens.fr/papers/astree_airbus_sas2007.pdf","https://www-cdn.anthropic.com/files/4zrzovbb/website/6867eadf760258fa8b42a504600cd3c036d16358.pdf"]},"facets":{},"traction":{"github_stars":null,"hn_points":14,"hn_comments":1,"hn_story_id":"50202217","hf_downloads":null,"hf_likes":null},"data_radar":null},"primary_evidence_page":{"is_primary":true,"source_match":true,"url":"https://www.anthropic.com/research/formal-verification","final_url":"https://www.anthropic.com/research/formal-verification","title":"Formal Verification","http_status":200,"content_type":"text/html; charset=utf-8","capture_method":"plain","fetched_at":"2026-10-23T00:02:39.901791+00:00","bytes":262401,"raw_path":"f6a301346a05e615cdbc2095a3fdc32b6b5a6109f77033291fc57d4e1fc9bb8d.html","content_hash":"925a3f2a45152be3257845c1c4c3e2f2335d8c968709311a0bdb6dc4f251b231","excerpt_chars":1200,"truncated":true,"excerpt":"LLMs are making formal software verification feasible at a much larger scale \\ Anthropic Skip to main content Chapter 01 : Introduction Keane Lucas, Aengus Lynch, Luna Tong, Luke Champine, Zygi Straznickas, Dino Distefano, Harshikaa Agrawal Evyatar Ben Asher, Ava Howell, Sami Jawhar, Chris Rohlf, Nickolai Zeldovich Nicholas Carlini, Rajashree Agrawal, Jason Gross Most software is riddled with bugs, which AI can now easily find and exploit . The standard defenses against cyberattacks are no longer sufficient: we need ways to prove that code is free of bugs. Formal software verification could provide this guarantee, but it has historically required years of expert effort, making it impractical for most real-world projects. In a collaboration with Theorem , however, we show this is no longer the case: LLMs can now produce formal proofs at large scale, enabling new software development workflows that include proven security guarantees. Using Claude, we proved security properties at the assembly code level (RISC-V) in over 100 real-world open-source software libraries. To our knowledge, this represents the first large-scale effort to use AI to apply formal verification to production..."},"evidence_pages":[{"is_primary":false,"source_match":false,"url":"https://sel4.systems/Research/pdfs/comprehensive-formal-verification-os-microkernel.pdf","final_url":"https://sel4.systems/Research/pdfs/comprehensive-formal-verification-os-microkernel.pdf","title":"Formal Verification","http_status":200,"content_type":"application/pdf","capture_method":"exa","fetched_at":"2026-10-23T00:03:18.663+00:00","bytes":1861193,"raw_path":"4aa2d86e6e029dc884b6e3b86ee272aa594ae115877bac6e9c90add32563d09f.pdf","content_hash":"a7180c9cf6a5c8da455fdf3809e0fc8b8bbfb15f68c2e44ad74b676bf6b3bbb3","excerpt_chars":1200,"truncated":true,"excerpt":"Comprehensive Formal Verification of an OS Microkernel GERWIN KLEIN, JUNE ANDRONICK, KEVIN ELPHINSTONE, TOBY MURRAY, THOMAS SEWELL, RAFAL KOLANSKI, and GERNOT HEISER, NICTA and UNSW, Sydney, Australia We present an in-depth coverage of the comprehensive machine-checked formal verification of seL4, a general-purpose operating system microkernel. We discuss the kernel design we used to make its verification tractable. We then describe the functional correctness proof of the kernel’s C implementation and we cover further steps that transform this result into a comprehensive formal verification of the kernel: a formally verified IPC fastpath, a proof that the binary code of the kernel correctly implements the C semantics, a proof of correct access-control enforcement, a proof of information-flow noninterference, a sound worst-case execution time analysis of the binary, and an automatic initialiser for user-level systems that connects kernel-level access-control enforcement with reasoning about system behaviour. We summarise these results and show how they integrate to form a coherent overall analysis, backed by machine-checked, end-to-end theorems. The seL4 microkernel is currently..."},{"is_primary":false,"source_match":false,"url":"https://www.astree.ens.fr/papers/astree_airbus_sas2007.pdf","final_url":"https://www.astree.ens.fr/papers/astree_airbus_sas2007.pdf","title":"Formal Verification","http_status":200,"content_type":"application/pdf","capture_method":"exa","fetched_at":"2026-10-23T00:03:18.629+00:00","bytes":69422,"raw_path":"43f736e7375148daa55e95da4a9a907e30b98c2cf246c8348dd6ce5247c966eb.pdf","content_hash":"ac858e322ff49d95e8e057ceef251dfad0cc71d3c1ca1c37e8f08b2ac846a956","excerpt_chars":1200,"truncated":true,"excerpt":"Astrée: from Research to Industry David Delmas and Jean Souyris Airbus France S.A.S. 316, route de Bayonne 31060 TOULOUSE Cedex 9, France {David.Delmas, Jean.Souyris}@Airbus.com Abstract. Airbus has started introducing abstract interpretation based static analysers into the verification process of some of its avionics software products. Industrial constraints require any such tool to be extremely precise, which can only be achieved after a twofold specialisation process: first, it must be de signed to verify a class of properties for a family of programs efficiently; sec ond, it must be parametric enough for the user to be able to fine tune the analy sis of any particular program of the family. This implies a close cooperation between the tool-providers and the end-users. Astrée is such a static analyser: it produces only a small number of false alarms when attempting to prove the ab sence of run-time errors in control/command programs written in C, and pro vides the user with enough options and directives to help reduce this number down to zero. Its specialisation process has been reported in several scientific papers, such as [1] and [2]. Through the description of analyses..."},{"is_primary":false,"source_match":false,"url":"https://www-cdn.anthropic.com/files/4zrzovbb/website/6867eadf760258fa8b42a504600cd3c036d16358.pdf","final_url":"https://www-cdn.anthropic.com/files/4zrzovbb/website/6867eadf760258fa8b42a504600cd3c036d16358.pdf","title":"Formal Verification","http_status":200,"content_type":"application/pdf","capture_method":"exa","fetched_at":"2026-10-23T00:03:18.625+00:00","bytes":200376,"raw_path":"9234b3d489e90dac0d0d9caa5b9dad04339da026735fa7e02e4368acc2b4d750.pdf","content_hash":"5f6c6c3f4c086d6d7b7106e914af768bf27cadaad2267c8d11196662b5ff1f6a","excerpt_chars":1200,"truncated":true,"excerpt":"Appendix: LLMs are making formal software verification feasible at a much larger scale October 22, 2026 What our OSS-Fuzz proofs prove 1 Building a Python interpreter in Rust 2 Details on Claude’s generated proofs 3 How do Claude’s proofs look to a human formal verification expert? 4 Details on the formal verification tasks we compare LLMs on 5 Assertions in real OSS-Fuzz entry points 5 What our OSS-Fuzz proofs prove Our proofs concern RISC-V binaries running on Linux. This gives us a stronger guarantee than if we had translated the fuzz tests into Rocq (which risks a translation error) or interpreted their source code (which risks our interpreter disagreeing with the actual behavior of the compiled binary). Instead, we interpret the compiled binary directly, using the formal semantics of RISC-V and Linux processes. We then define the notion of a crash: For our purposes, a crash means a fault (e.g. an unmapped memory access), an abort (e.g. a failed assert), or any unmodeled behavior (e.g. a syscall that isn’t defined in our Linux process semantics). The basic theorem statement that we aim to prove is that no possible input to a given binary results in a crash. If we can prove..."}],"related_signals":[{"id":"7d6ec487-6763-48f8-bd14-21c7544abca4","url":"https://onlylabs.fyi/signals/7d6ec487-6763-48f8-bd14-21c7544abca4","source_url":"https://www.anthropic.com/research/zero-days","title":"Zero Days","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"anthropic","name":"Anthropic","category":"frontier-lab"},"occurred_at":"2026-02-05T00:00:00.000Z","first_seen_at":"2026-06-17T20:00:31.646229+00:00","date_source":"page.visible_date"},{"id":"f6ee7530-aad8-444f-a772-290b240701ff","url":"https://onlylabs.fyi/signals/f6ee7530-aad8-444f-a772-290b240701ff","source_url":"https://www.anthropic.com/research/property-based-testing","title":"Property Based Testing","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"anthropic","name":"Anthropic","category":"frontier-lab"},"occurred_at":"2026-01-14T00:00:00.000Z","first_seen_at":"2026-06-17T20:00:31.646229+00:00","date_source":"page.visible_date"},{"id":"b18d4494-1bde-4128-800c-592e37b2bb51","url":"https://onlylabs.fyi/signals/b18d4494-1bde-4128-800c-592e37b2bb51","source_url":"https://www.anthropic.com/research/project-vend-2","title":"Project Vend 2","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"anthropic","name":"Anthropic","category":"frontier-lab"},"occurred_at":"2025-12-18T00:00:00.000Z","first_seen_at":"2026-06-09T02:17:26.339488+00:00","date_source":"page.visible_date"},{"id":"dcc9f140-6412-4147-aa53-118fd4e02404","url":"https://onlylabs.fyi/signals/dcc9f140-6412-4147-aa53-118fd4e02404","source_url":"https://www.anthropic.com/research/smart-contracts","title":"Smart Contracts","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"anthropic","name":"Anthropic","category":"frontier-lab"},"occurred_at":"2025-12-01T00:00:00.000Z","first_seen_at":"2026-06-17T20:00:31.646229+00:00","date_source":"page.visible_date"},{"id":"b5ffd9d4-bdf4-45ae-9386-86fa6dee7a81","url":"https://onlylabs.fyi/signals/b5ffd9d4-bdf4-45ae-9386-86fa6dee7a81","source_url":"https://www.anthropic.com/research/nuclear-safeguards-for-ai","title":"Nuclear Safeguards For Ai","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"anthropic","name":"Anthropic","category":"frontier-lab"},"occurred_at":"2025-08-21T00:00:00.000Z","first_seen_at":"2026-06-17T20:00:31.646229+00:00","date_source":"page.visible_date"},{"id":"271549ea-a84e-415a-9d8c-3a18ead586df","url":"https://onlylabs.fyi/signals/271549ea-a84e-415a-9d8c-3a18ead586df","source_url":"https://www.anthropic.com/research/cyber-toolkits","title":"Cyber Toolkits","context":null,"kind":{"key":"post_published","label":"Writing"},"org":{"slug":"anthropic","name":"Anthropic","category":"frontier-lab"},"occurred_at":"2025-06-13T00:00:00.000Z","first_seen_at":"2026-06-17T20:00:31.646229+00:00","date_source":"page.visible_date"}]}