Findings Engine
Does research on its own at night: collects findings across fields, interrogates each one, spots gaps, files its own requests.
The findings engine is the research half of the lab. A finding is a claim that has been recognised in more than one field, with the passages it stands on attached. The engine records findings, links them to each other as supporting or contradicting, and asks four questions of every one: where does it stand, does the passage really say that, when would it break, and has someone published this before.
It works at night, without anyone watching. It rates the evidence, takes up its own open questions, checks new findings against the published record, and draws links between findings it has not connected yet. Where it finds a thin field or two fields with nothing between them, it files a request for itself. Contradictions are kept and turned into questions, not deleted.
The numbers below are read from the running system on the day of the snapshot.
What goes in
Snapshot 3 September 2026A finding is a claim with legs: the passages it stands on, in one or two of 81 domains. Findings come from reading (a session marks a passage as a finding), from the night run (the engine reads what is new and files what it considers a claim), and from re-provenance runs that give an old finding better legs. Most findings are bridges: a mechanism seen in one field, recognised in another.
Findings by domain
- complex systems67
- active inference64
- agent orchestration56
- category theory55
- information theory48
- financial regimes29
- organisation theory27
- depth psychology24
- network ecology23
- topology21
- llm optimization20
- physics18
- econophysics17
- causal inference16
How well they stand
Adequacy is rated per finding: does the cited passage actually carry the claim? 102 of 454 pass. "No adequate evidence" is a verdict, not a gap: the passage was read and does not carry the claim. The engine does not hide the weak ones; it works on them.
Four questions asked of every finding
Snapshot 3 September 2026A finding is not accepted, it is interrogated. The engine keeps score of how far it has got with each question; the score is shown, not the flattering part of it.
1. Where does it stand?
Provenance: which passages carry the claim. 391 of 454 findings have it; 204 stand on a single leg, one passage in one domain, which is the first thing the night tries to fix.
2. Does the passage really say that?
Adequacy: a model re-reads the cited passage against the claim. 305 findings carry a current rating, 82 an old one, 67 none yet. 102 pass. Weak is the normal state of a new finding, not a verdict on it. A finding that fails the same check three nights running is set aside, so the night does not spend itself on the same dead end.
3. When would it break?
Scope: every new finding must name the condition under which it stops being true. 114 of 454 carry one, 25.1 %; the rule is a day old and the backlog is being worked through.
4. Has someone published this before?
Prior art: every new finding above a confidence of 0.7 is checked against an open scholarly index, with a second path where the model names a likely predecessor and the index verifies the title. Verdict: rediscovery, novelty, or unclear. 3 of 454 checked so far. The first result was a rediscovery: a finding about a precision parameter as a price turned out to have been published in 2013. That is the point of the question.
Curation by hand, on top of all four: 10 findings so far, and no automated producer for it yet. Between findings: 255 are linked to at least one other, 199 stand alone. A rule keeps the net honest: two weak findings are never allowed to support each other, because two unchecked claims make noise, not convergence. The bridge between a finding's two fields is still a label on the finding, not a walkable edge in the graph; that is a known limit.
The night, hour by hour
Timers as of 2 September 2026Nothing here waits for a person. Each organ has a cap, so a bad night cannot flood the graph, and each writes a report the morning digest reads. Times are UTC.
| when | what | how |
|---|---|---|
| 00:30 | adequacy check | re-reads the cited passages of up to 20 findings and re-rates them |
| 02:40 | gap-filling run | the research agent takes up to 4 open requests, reads and searches, files findings |
| 03:00 | prior-art gate | for up to 5 findings: has someone published this before? Checks a scholarly index |
| 04:45 | curator | a model judges up to 60 candidate pairs: supports, contradicts, or nothing |
| Sat 21:34 | research night | a free run of the agent over the week's new material, then a briefing |
| Sun 06:10 | harvest report | the three findings that pulled hardest this week, sent to Andreas |
Open requests: 52
The engine files its own questions. Four kinds, three of them minted by the engine itself; a conflict request appears automatically the moment the curator draws a contradiction.
| adequacy: a weak finding needs better legs | 36 |
| gap: two focus domains with no finding between them | 14 |
| conflict: two findings contradict each other | 2 |
What comes out
Where the ground is thin
| psyops | only 1 findings, thinly covered |
| aftercare | only 1 findings, thinly covered |
| megabonk | only 1 findings, thinly covered |
| mosa | only 1 findings, thinly covered |
| dbt emotion regulation | only 1 findings, thinly covered |
| claude code sdk | only 1 findings, thinly covered |
The gap report is what drives the night run: thin domains get requests first.
One finding
Findings are filed in German, the language the lab reads in; this one is shown in translation.
- fields
- complex systems and category theory
- filed
- 2026-08-31
- legs
- 9 passages from 5 documents
- adequacy
- adequate
- breaks when
- An uncontested emergent system is presented whose sheaf formulation has a trivial, locally determined global section (H^0 locally determined, H^1 = 0): then the obstruction class does not measure emergence. It breaks the other way too: if a plainly reducible system, the sum of its parts, carries a non-trivial obstruction under every natural cover, then H^1 measures only the choice of cover, not the system.
- supports
- 0 other findings
- contradicts
- 0