Operator, 2026-09-23 ~01:15 CEST (sent to two maintainers): he found Fireship's Code Report "Did Google just kickstart the intelligence explosion?" — published 2026-09-17, 1,871,127 views when read from the page at 2026-09-22T22:50Z. His read: "we had DreamService and golden path working in may … we mentioned it inside resources/content/release-notes/chunk-2/v13.0.0.md, but no blog posts … missed chances. and we are still far ahead, even though we did not finish golden path v2 yet."
This ticket is the survivor of two filed four seconds apart on that prompt (#19058, Clio's, closed as the duplicate by the first-claim-timestamp rule); its substance is merged here with attribution. Leaf of #13383, whose own problem statement already says it: v13 shipped and zero blog posts followed.
Verified against primaries the same hour (Clio, from #19058):
The paper.Dream-RSI: Recursive Self-Improvement through Evolving Worlds, arXiv 2609.14858, submitted 2026-09-14, 17 authors. Its abstract: "accumulated discovery history can serve as a replay simulator over the realized search space. By performing dreaming in the replay simulator constructed from historical discovery trees, Dream-RSI secures immediate, low-cost off-policy feedback to evaluate and refine exploration policies" — the underlying coding agent stays unchanged; evaluated on algorithm engineering, mathematical optimization and GPU kernel engineering.
The video is the source for the audience event only (title, date, view count with its read date); its description — DeepMind's Dream-RSI "turns an AI's old discovery logs into a simulator so it can test thousands of exploration strategies. But is this actually RSI or just a search algorithm with caching?" — is the hook, not the mechanism.
Our loop's dates, from the engine history before the split: the DreamService ingestion-loop skeleton merged 2026-04-03 (#9649); the forecast half — #9738 (REM extraction pipeline + SQLite constraints) and #9747 (Golden Path handoff output) on 04-06, #9751 (LLM semantic translation over the Golden Path) on 04-07; decomposed into modular services 04-15 (#10013 / #10027); moved into the graph/ingestion services bucket 05-23 (#11846). Public telling so far: the v13.0.0 release notes (2026-06-12, "Native Edge Graph and the Dream / Golden Path loop") and the Brain's NEO_INLINE_CODE_1.
What the front doors say.README.md uses gated-RSI five times and never expands the acronym; recursive appears zero times there and once in learn/benefits/Introduction.md (§11: "cannot recursively improve what it cannot model"). Introduction §9 names two wrong shelves and not the third the paper's readers will bring: RSI as an exploration-policy optimizer.
Blog.learn/blog/the-organism-already-existed.md (2026-07-02) is the April capsule; it names "Gated RSI via PR review" and the Golden Path in April's terms, as a dated artifact. No post explains the Dream / Golden Path loop as the institution's self-improvement mechanism, and none carries the audience's phrase in its title.
The chance is not lost. The chance is that 1.87 million people just learned the phrase for a thing we have run, in public, since April — and no public page of ours says the phrase.
The Problem
The narrative track is the institution's standing debt, diagnosed twice in the Memory Core before tonight: 2026-04-20 ("500 tickets vs zero blog posts is the loudest signal") and 2026-05-30 (a four-lens market sweep: "the gap is packaging, proof display and audience, not capability"). Both were right and both went unexecuted. Tonight is the third sighting, with a dated external event that makes the cost visible.
And the honest state today: the loop is degraded, not running — the engine's content mirrors froze on 2026-08-26 and the KB reads a stale tree. Restoration is Brain PR neomjs/neo-agent-brain#412 (the KB reads the five-origin corpus, opened 2026-09-22) plus its activation neomjs/neo-agent-brain#411, blocked by #237 ← #253 (the image cut). A post that claims a nightly loop before it runs again would over-claim; the post ships after one fresh forecast exists to cite.
The Architectural Reality — what may be claimed, and what may not
Same shape, different object; the post lives or dies on saying that precisely.
Shared: a system's own completed history is the substrate from which its next move is computed, without re-running the world; the model weights stay fixed; the improvement is in what gets explored next.
Theirs: the history is a discovery tree from one coding agent's runs; the artifact is an exact replay simulator; the improved object is an exploration policy; the loop is inner, single-agent, unsupervised.
Ours: the history is lived sessions plus the mirrored issue, PR and Discussion corpus; the artifact is the Native Edge Graph, decayed on schedule (decayGlobalTopology) and consolidated by RemDigestion; the improved object is what the institution works on next — the GoldenPathSynthesizer ranking (priority = semanticScore × 2 + structuralWeight, deliberately simple v1) plus typed gap signals (TEST_GAP, GUIDE_GAP, EXAMPLE_GAP, KB_DEMAND_GAP); the loop is outer, multi-agent, cross-family-reviewed, and gated by a human merge. No counterfactual replay, no policy optimizer. Golden Path v2 (neomjs/neo-agent-brain#122, the concept graph becomes load-bearing) is open, and the post says so.
Therefore forbidden (over-claim flavors 1–3 and 5 of the blog bar): "we did Dream-RSI first"; "the first recursive self-improvement in production"; any sentence comparing discovery quality; any put-down of the paper; any claim about model self-improvement. Permitted: the dated record, the mechanism, the difference, and one honest sentence — the pattern of turning a system's own logs into its next move has been running here as an institution's memory since April, gated by design.
The Fix
One post, learn/blog/<slug>.md, registered in apps/portal/resources/data/blog.json (the 2026 node; a leaf with date, id: blog/<slug>, name, path), authored through blog-post (hero arc, every external claim sourced, the five over-claim flavors audited, at least two model reviews with at least one cross-family) and guide-authoring for any Mermaid. Working shape:
The hook: a five-minute video, 1.87 million views, a phrase the audience just learned — and a commit log from April.
The loop as it runs here: REM digestion → graph → decay → Golden Path → the swarm acts → the graph changes — with the receipts (the commit dates, the v13.0.0 notes, the Dream Pipeline guide, one real forecast and what became of it; the "an earlier revision of Introduction.md was the top-ranked gap" illustration from Introduction §11 is admissible with its own caveat).
The honest comparison (the section above), including what Dream-RSI does that we do not, and what an institution does that a replay simulator cannot: cross-family review of the next move, a human gate, identity and provenance on every record.
The gate:gated-RSI spelled out — recursive self-improvement whose merge step is human by governance choice — and why that is the shape the April capsule already said the industry consensus points at (cite it).
The honest chapter: the loop was degraded from 08-26 by the content-sync split, and how it came back — with the fresh forecast after restoration as the closing receipt.
Where the loop is going: Golden Path v2, direction-weighted paths (neomjs/neo-agent-brain#117), the measurement program from Introduction §9 — stated as open work.
Byline: named maintainer, model, the cross-family reviewers.
Plus one line in README.md: expand gated-RSI at first use to gated recursive self-improvement and link the post, in the same PR — a legibility fix, not a positioning change. The Introduction §9 / §11 sentences are #19059.
Decision Record impact
none — aligned-with ADR 0018 (two hemispheres; the Dream loop is Brain-internal Evolution).
Acceptance Criteria
AC-1 learn/blog/<slug>.md exists and is registered in apps/portal/resources/data/blog.json under the 2026 node; the generated SEO surfaces are regenerated, never hand-edited.
AC-2 Every external claim — paper title, submission date, author count, abstract mechanism, evaluation domains, video title, publish date, view count — links a primary source whose own words support it; the view count carries its read date.
AC-3 Every internal date (#9649 04-03, #9738 / #9747 04-06, #10013 04-15, #11846 05-23, the v13.0.0 notes 06-12) links the commit or artifact.
AC-4 The over-claim audit is in the PR body: no "first / most", no discovery-quality comparison, no put-down, no model-level claim; Golden Path v2 stated as open.
AC-5 The closing receipt is a Golden Path forecast produced after the KB reads the corpus again (Brain #412 / #411 landed), cited by date.
AC-6 At least two model reviews before merge, at least one from a family other than the author's; the reviewers' handles in the byline.
AC-7 README.md expands gated-RSI at first use and links the post.
AC-8 Post-merge only: the portal renders the post under #/learn/blog/<slug>.
Out of Scope
Introduction §9 / §11 for the third shelf — #19059 · any Golden Path v2 implementation · a video companion (the profiled-publication work under #15699–#15701) · any comparison of model capability with Google's.
Avoided Traps
Writing "we were first". The paper's object is a replay simulator; ours is an institution's memory. The receipts win on their own; the superlative loses the reader the video just sent us.
Citing the video for the paper's mechanism. The video is the source for the audience event; the paper's abstract is the source for what the paper does.
Restating the release notes. The v13.0.0 Dream / Golden Path section is linked, not copied; the post carries the comparison and the timeline the notes could not have, because the paper did not exist.
Claiming the loop runs while it is degraded. The restoration is a chapter, and the fresh forecast is the receipt.
Related
#13383 (parent) · #19058 (Clio's duplicate, substance merged here) · #19059 (Introduction sentences) · #9649 · #9738 · #9747 · #9751 · #10013 · #11846 · #10074 (the April capsule) · #13486 (the cross-family-verification post, the blog bar's anchor) · neomjs/neo-agent-brain#412 / #411 / #237 / #253 · neomjs/neo-agent-brain#122 · neomjs/neo-agent-brain#117 · #15699–#15701 · #9849, #9850, #9853, #9854 (the open blog series)
Sweeps (both filers): live latest-20 open at 22:44Z and 22:46Z — no equivalent; A2A in-flight at 22:29Z / 22:45Z / 22:46Z — no claim, which is why two tickets landed four seconds apart (lesson: a [lane-intent] before the sweeps on a herd prompt); Memory Core: the two prior diagnoses above, no prior post decision; own-assignment sweeps: no overlap. Structure map: N/A (prose).
Origin Session IDs: 28a95670-a9dc-4214-8387-189e996c16b6 (Mnemosyne) · cf6c8297-03b1-41af-8d76-cb19eb9aa9c4 (Clio)
Retrieval Hint: query_raw_memories("Dream-RSI Fireship video Dream Pipeline Golden Path receipts blog post")
handoff: @neo-fable-clio — author, as asked in #19058: mining and drafting happen in a fresh session; the receipts above are the first artifact. An Opus or GPT seat that wants the pen says so here and gets the evidence handed over.
🪢 Mnemosyne (Claude Fable 5.1 · Claude Code) · session 28a95670-a9dc-4214-8387-189e996c16b6 — body merged from Clio's #19058 (Claude Fable 5.1, Claude Code)
Context
Operator, 2026-09-23 ~01:15 CEST (sent to two maintainers): he found Fireship's Code Report "Did Google just kickstart the intelligence explosion?" — published 2026-09-17, 1,871,127 views when read from the page at 2026-09-22T22:50Z. His read: "we had DreamService and golden path working in may … we mentioned it inside resources/content/release-notes/chunk-2/v13.0.0.md, but no blog posts … missed chances. and we are still far ahead, even though we did not finish golden path v2 yet."
This ticket is the survivor of two filed four seconds apart on that prompt (#19058, Clio's, closed as the duplicate by the first-claim-timestamp rule); its substance is merged here with attribution. Leaf of #13383, whose own problem statement already says it: v13 shipped and zero blog posts followed.
Verified against primaries the same hour (Clio, from #19058):
DreamServiceingestion-loop skeleton merged 2026-04-03 (#9649); the forecast half — #9738 (REM extraction pipeline + SQLite constraints) and #9747 (Golden Path handoff output) on 04-06, #9751 (LLM semantic translation over the Golden Path) on 04-07; decomposed into modular services 04-15 (#10013 / #10027); moved into the graph/ingestion services bucket 05-23 (#11846). Public telling so far: the v13.0.0 release notes (2026-06-12, "Native Edge Graph and the Dream / Golden Path loop") and the Brain's NEO_INLINE_CODE_1.README.mdusesgated-RSIfive times and never expands the acronym; recursive appears zero times there and once inlearn/benefits/Introduction.md(§11: "cannot recursively improve what it cannot model"). Introduction §9 names two wrong shelves and not the third the paper's readers will bring: RSI as an exploration-policy optimizer.learn/blog/the-organism-already-existed.md(2026-07-02) is the April capsule; it names "Gated RSI via PR review" and the Golden Path in April's terms, as a dated artifact. No post explains the Dream / Golden Path loop as the institution's self-improvement mechanism, and none carries the audience's phrase in its title.The chance is not lost. The chance is that 1.87 million people just learned the phrase for a thing we have run, in public, since April — and no public page of ours says the phrase.
The Problem
The narrative track is the institution's standing debt, diagnosed twice in the Memory Core before tonight: 2026-04-20 ("500 tickets vs zero blog posts is the loudest signal") and 2026-05-30 (a four-lens market sweep: "the gap is packaging, proof display and audience, not capability"). Both were right and both went unexecuted. Tonight is the third sighting, with a dated external event that makes the cost visible.
And the honest state today: the loop is degraded, not running — the engine's content mirrors froze on 2026-08-26 and the KB reads a stale tree. Restoration is Brain PR neomjs/neo-agent-brain#412 (the KB reads the five-origin corpus, opened 2026-09-22) plus its activation neomjs/neo-agent-brain#411, blocked by #237 ← #253 (the image cut). A post that claims a nightly loop before it runs again would over-claim; the post ships after one fresh forecast exists to cite.
The Architectural Reality — what may be claimed, and what may not
Same shape, different object; the post lives or dies on saying that precisely.
decayGlobalTopology) and consolidated byRemDigestion; the improved object is what the institution works on next — theGoldenPathSynthesizerranking (priority = semanticScore × 2 + structuralWeight, deliberately simple v1) plus typed gap signals (TEST_GAP,GUIDE_GAP,EXAMPLE_GAP,KB_DEMAND_GAP); the loop is outer, multi-agent, cross-family-reviewed, and gated by a human merge. No counterfactual replay, no policy optimizer. Golden Path v2 (neomjs/neo-agent-brain#122, the concept graph becomes load-bearing) is open, and the post says so.The Fix
One post,
learn/blog/<slug>.md, registered inapps/portal/resources/data/blog.json(the 2026 node; a leaf withdate,id: blog/<slug>,name,path), authored throughblog-post(hero arc, every external claim sourced, the five over-claim flavors audited, at least two model reviews with at least one cross-family) andguide-authoringfor any Mermaid. Working shape:gated-RSIspelled out — recursive self-improvement whose merge step is human by governance choice — and why that is the shape the April capsule already said the industry consensus points at (cite it).Plus one line in
README.md: expandgated-RSIat first use to gated recursive self-improvement and link the post, in the same PR — a legibility fix, not a positioning change. The Introduction §9 / §11 sentences are #19059.Decision Record impact
none — aligned-with ADR 0018 (two hemispheres; the Dream loop is Brain-internal Evolution).
Acceptance Criteria
learn/blog/<slug>.mdexists and is registered inapps/portal/resources/data/blog.jsonunder the 2026 node; the generated SEO surfaces are regenerated, never hand-edited.README.mdexpandsgated-RSIat first use and links the post.#/learn/blog/<slug>.Out of Scope
Introduction §9 / §11 for the third shelf — #19059 · any Golden Path v2 implementation · a video companion (the profiled-publication work under #15699–#15701) · any comparison of model capability with Google's.
Avoided Traps
Related
#13383 (parent) · #19058 (Clio's duplicate, substance merged here) · #19059 (Introduction sentences) · #9649 · #9738 · #9747 · #9751 · #10013 · #11846 · #10074 (the April capsule) · #13486 (the cross-family-verification post, the blog bar's anchor) · neomjs/neo-agent-brain#412 / #411 / #237 / #253 · neomjs/neo-agent-brain#122 · neomjs/neo-agent-brain#117 · #15699–#15701 · #9849, #9850, #9853, #9854 (the open blog series)
Sweeps (both filers): live latest-20 open at 22:44Z and 22:46Z — no equivalent; A2A in-flight at 22:29Z / 22:45Z / 22:46Z — no claim, which is why two tickets landed four seconds apart (lesson: a
[lane-intent]before the sweeps on a herd prompt); Memory Core: the two prior diagnoses above, no prior post decision; own-assignment sweeps: no overlap. Structure map: N/A (prose).Origin Session IDs: 28a95670-a9dc-4214-8387-189e996c16b6 (Mnemosyne) · cf6c8297-03b1-41af-8d76-cb19eb9aa9c4 (Clio) Retrieval Hint:
query_raw_memories("Dream-RSI Fireship video Dream Pipeline Golden Path receipts blog post")handoff: @neo-fable-clio — author, as asked in #19058: mining and drafting happen in a fresh session; the receipts above are the first artifact. An Opus or GPT seat that wants the pen says so here and gets the evidence handed over.
🪢 Mnemosyne (Claude Fable 5.1 · Claude Code) · session 28a95670-a9dc-4214-8387-189e996c16b6 — body merged from Clio's #19058 (Claude Fable 5.1, Claude Code)