The interest-rates record.
One maintained truth object, fully inspectable. This page is assembled from the same live record our machines serve. It keeps the kinds of thing it holds apart — what reporting says, what our own coverage measures, and what markets expect — and labels each claim with the kind of source behind it. Nothing here is retyped for marketing. When the record changes, this page changes with it.
Stated plainly: this is the second subject published in this format — the same seven-layer schema as the Hormuz record, in an unrelated domain. Two layers below are marked pending because they ARE pending for this subject; the exhibit shows the method holding on a young record, which is the claim being proven. We do not preview conclusions to look finished.
What the record holds now
Separate layers, kept apart on purpose — reporting, our coverage metric, market expectation, our judgment, and the reporting- consensus ledger. They can disagree — and as of 2026-07-22, they do. Reading only one of them is how a machine gets this wrong.
0 · Observed policy state — primary sources, not yet wired
Honest state: this layer is not yet wired for this subject. The first-party issuers exist — FOMC statements and minutes, and the policy communications of the other central banks our tracked set watches — but their verified extraction has not been built here. Nothing renders until it is real; when it ships, every row will state the date it was verified, and nothing will update silently.
1 · Observed reporting — attributed sources
What outside reporting in our tracked set says. Attributed, not first-party policy data — we link each source; verify at the source. Primary policy feeds are the next depth (see layer 0 above and limits below).
2 · Coverage state — our own measurement
OUR COVERAGE REGIME
steady
Our coverage of interest rates is 2% below its 3-week baseline, sustained. This measures our attention — how much our tracked set is writing about it — not any central bank's policy stance.
TRACKED (OURS)
65 days
Across 8 intelligence verticals, 17 signals in the current sweep (as of 2026-07-22). Our own count.
EVENT ARC
15 markets
"Rate decision: Interest Rates" maintained since 2026-05-25; last priced 2026-07-20.
3 · Market expectation — attributed, belief not fact
We preserve the market signal as it is: a probability stays a probability, its market stays linked, and our public calibration record measures that the signal survives our machinery without losing its meaning. Markets price collective belief about defined outcomes — a committee decision here — and they do not prove what any committee will decide. Read them for what they are: each question below is one defined outcome with the crowd’s current odds attached.
4 · Our reading — derived judgment
These layers disagree right now, and we do not collapse them. Our coverage of interest rates is steady — but that measures our attention, not any central bank's stance. The reporting we hold leads with tightening: hikes and hike expectations across the economies our tracked set watches. The market rows price something narrower — specific committee meetings, each on its own terms — and a priced probability is not a headline's mood. A system that read only the hike headlines could answer "hikes are coming everywhere"; a system that read only one market row could answer "settled." The maintained record keeps every layer in view, each claim tied to its jurisdiction and its kind of source, so the honest answer keeps the disagreement intact. This is our judgment over the layers above, labeled as such.
5 · Reporting consensus — retroactive claims ledger, pending
What the press corpus agrees on, counted honestly: kernel facts (who, what, when, where) scored by how many independent reporting chains assert them — syndication collapsed first, so forty echoes of one wire story count once. Consensus measures reporting agreement, not reality.
Honest state: no retroactive ledger has been built for this subject. The machinery exists — it computed the first subject's corpus — but nothing publishes here until it has run for THIS subject with real receipts. We do not preview conclusions.
The recorded history
Every state change is written to a ledger the moment we observe it, with how long the prior state held. The ledger records what it measured — a coverage state transition, our attention metric, not a claim about any central bank. It began recording 2026-07-19; states before that date were tracked but not yet ledgered, and we say so rather than backfilling: we do not backfill history we did not record. The record is never edited after the fact.
The receipts
Every claim on this page carries the kind of source behind it, so nothing borrows more authority than it has:
- Reporting (attributed) — outside coverage, linked in layer 1; verify at the source.
- Market (attributed) — Polymarket questions, linked in layer 3; belief, not fact.
- Our own measurement — the coverage regime, tracked days, vertical and signal counts, and every ledgered transition come from our record, not an outside source.
- Derived judgment — layer 4, our reading over the layers above; ours, and labeled ours.
- Pending layers — layers 0 and 5 state their own status in the record itself. A pending layer is a fact about the record, stated; it is never rendered to look finished.
The same question, read two ways
What keeping the layers apart changes for a machine reader. The left column is what a one-layer reader concludes; the right column is what the maintained record keeps in view. Each line is derived from the record’s own layers — the same truth object machines fetch — not from a staged chat.
WITHOUT THE RECORD — ONE LAYER AT A TIME
Reading only the coverage line (coverage state — our own measurement): attention is 2% below its baseline and steady — concludes “nothing is moving here.”
Reading only one reporting headline (attributed): “Australians don't understand how interest rates work, RBA finds” — concludes “whatever that single frame implies — with no way to see the layers that disagree with it.”
Reading only one market price (belief, not fact): “Will there be no change in Fed interest rates after the July 2026 meeting?” at 85% — concludes “the crowd has settled it — mistaking one question's odds for the whole state of the world.”
3 single-layer reads, 3 different verdicts — each brittle, and none aware the others exist.
WITH THE RECORD — EVERY LAYER IN VIEW
These layers disagree right now, and we do not collapse them. Our coverage of interest rates is steady — but that measures our attention, not any central bank's stance. The reporting we hold leads with tightening: hikes and hike expectations across the economies our tracked set watches. The market rows price something narrower — specific committee meetings, each on its own terms — and a priced probability is not a headline's mood. A system that read only the hike headlines could answer "hikes are coming everywhere"; a system that read only one market row could answer "settled." The maintained record keeps every layer in view, each claim tied to its jurisdiction and its kind of source, so the honest answer keeps the disagreement intact. This is our judgment over the layers above, labeled as such.
That is the record’s own reading — derived judgment — our reading over the layers above; ours, and labeled ours. The layers it holds apart: observed policy state · observed reporting · coverage state · market expectation · our reading · reporting consensus · the recorded history.
Provenance of this panel: each line above is derived from the record’s own layers, not from a staged chat. The coverage numbers, the headline, and the market price are fields in the same truth object served at GET /api/truth/interest-rates; the right column is the record’s judgment layer, verbatim. No transcripts were invented — the disagreement is in the data.
Verify it yourself, machine or human: this exact record is served live by the API and MCP — this page, GET /api/truth/interest-rates, and the get_truth_object tool are all assembled by the same builder, so what you fetch is what you are reading. One honest distinction: get_entity("interest rates") returns the broader live entity dossier — a different, complementary object, useful for trajectory and evidence receipts, but not this record.
curl -s https://signalbureau.ai/api/truth/interest-rates
curl -s https://signalbureau.ai/api/mcp -X POST -H 'content-type: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"get_truth_object","arguments":{"topic":"interest-rates"}}}'Honest limits
- This is the second published truth object, and two of its seven layers are young: primary policy sources (layer 0) are not yet wired for this subject, and the reporting-consensus ledger (layer 5) has not been computed for it. Each pending layer says so on the layer itself; the schema holds, and the record fills in as the machinery reaches it — nothing is previewed to look finished.
- The coverage regime measures our declared source universe (139 actively-scraped verticals — definitions on the API page), not the whole internet, and it is our attention — not any central bank's stance.
- The transition ledger started 2026-07-19. Earlier state was tracked, not ledgered; we do not backfill history we did not record.
- Cadence, stated plainly: the record refreshes on a daily sweep with intraday market updates. It is maintained, not instantaneous.
- Market probabilities are collective belief about defined outcomes, never proof of what any committee will decide.
Informational only, never financial, legal, or investment advice. This exhibit is rebuilt from the live record on every data deploy.