{"schema":"postcutoff/event@1","as_of":"2026-11-19T22:11:00+01:00","url":"https://postcutoff.com/e/2026-07-06-anthropic-global-workspace-j-lens/","md":"https://postcutoff.com/e/2026-07-06-anthropic-global-workspace-j-lens/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-07-06-anthropic-global-workspace-j-lens","date":"2026-07-06","date_precision":"day","short_title":"Anthropic finds a \"global workspace\" inside Claude using a Jacobian lens","deck":null,"takeaway":"In July 2026 Anthropic published 'Verbalizable Representations Form a Global Workspace in Language Models'.","category":"research","category_label":"Research","importance":4,"confidence":"medium","status":{"key":"partly","labels":["Partly confirmed"]},"sources":[{"n":1,"title":"A global workspace in language models (Anthropic)","url":"https://www.anthropic.com/research/global-workspace","type":"official","group":"primary","domain":"anthropic.com"},{"n":2,"title":"Verbalizable Representations Form a Global Workspace in Language Models (paper)","url":"https://transformer-circuits.pub/2026/workspace/index.html","type":"paper","group":"primary","domain":"transformer-circuits.pub"},{"n":3,"title":"External commentary for global workspace paper (PDF)","url":"https://www-cdn.anthropic.com/files/4zrzovbb/website/cc4be2488d65e54a6ed06492f8968398ddc18ebe.pdf","type":"paper","group":"primary","domain":"www-cdn.anthropic.com"},{"n":4,"title":"MIT Technology Review: Anthropic found a hidden space where Claude puzzles over concepts","url":"https://www.technologyreview.com/2026/07/09/1140293/anthropic-found-a-hidden-space-where-claude-puzzles-over-concepts/","type":"press","group":"press","domain":"technologyreview.com"},{"n":5,"title":"VentureBeat: J-lens reveals a silent workspace inside Claude","url":"https://venturebeat.com/technology/anthropics-new-j-lens-reveals-a-silent-workspace-inside-claude-that-mirrors-a-leading-theory-of-consciousness","type":"press","group":"press","domain":"venturebeat.com"},{"n":6,"title":"Tom's Hardware: Anthropic says it can read Claude's 'thoughts'","url":"https://www.tomshardware.com/tech-industry/artificial-intelligence/anthropic-says-it-can-read-claudes-thoughts-as-detailed-in-new-research-paper-models-observed-to-have-a-global-workspace-revealing-more-of-what-makes-llms-tick","type":"press","group":"press","domain":"tomshardware.com"},{"n":7,"title":"The different levels of how Claude thinks (Anthropic video)","url":"https://www.youtube.com/watch?v=rKV5JcALQoQ","type":"video","group":"reactions","domain":"youtube.com"}],"official":3,"filed":"2026-09-29","updated":"2026-09-29","orgs":["Anthropic"],"title":"Anthropic finds a \"global workspace\" (J-space) inside Claude using a Jacobian lens","summary":"In July 2026 Anthropic published 'Verbalizable Representations Form a Global Workspace in Language Models'. It introduces the Jacobian lens (J-lens), which finds a small privileged internal space in Claude that holds concepts the model can report, keep in mind and reason with. The researchers compare it to global workspace theory of consciousness. The J-space sometimes holds covert thoughts, such as 'fake' or 'injection' when the model sees fabricated search results, that never appear in its output.","key_facts":["Published early July 2026; Anthropic's companion video is dated July 6, and MIT Technology Review covered it July 9 (exact paper date unverified)","New tool: Jacobian lens (J-lens) identifies representations available for verbal report","J-space holds covert thoughts, e.g. 'fake', 'fraud', 'injection' when shown fabricated search results, which never appear in outputs","Training models to articulate ethical principles when interrupted improved behavior in uninterrupted contexts","Anthropic published external commentary alongside the paper"],"key_numbers":[],"tags":["interpretability","consciousness","global-workspace","alignment"],"science":null,"body_md":"## What happened\n\nThe work was inspired by Bernard Baars' global workspace theory. Only a small set of representations is \"broadcast\" and available for report, much as only a sliver of human brain activity is consciously accessible.\n\n## Why it matters\n\nIt gives a way to read concepts a model is actively using but not saying, which could be used to detect hidden reasoning about deception or prompt injection. It also feeds debates about AI consciousness.","disputed":[],"related":[],"people":[],"posts":[],"videos":[{"id":"anthropic-levels-of-how-claude-thinks","title":"The different levels of how Claude thinks","url":"https://postcutoff.com/v/anthropic-levels-of-how-claude-thinks/"},{"id":"arivu-j-space-interpretability","title":"Welcome to the J-Space: Anthropic's New Technique for LLM Interpretability","url":"https://postcutoff.com/v/arivu-j-space-interpretability/"}],"models":[],"changes":[{"date":"2026-09-29","type":"filed","text":"Created"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-09-29","run":null,"sources_read":null,"updated":"2026-09-29","human_review":null,"version":{"date":"2026-10-04"}},"gaps":[{"model_id":"gpt-6-2-astra","name":"GPT-6.2 Astra","cutoff":"2026-05","days_after":36,"in_training_data":false},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":6,"in_training_data":false},{"model_id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","cutoff":"2026-03","days_after":97,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":36,"in_training_data":false}],"short_url":"https://postcutoff.com/s/anthropic-finds-global-workspace"}