A debt five editions old is paid. A promise three days old is not.

This page is the current briefing, and it always lives at empowered.vote/briefing. Every edition also keeps a permanent address of its own, so a number you read here can be checked against the date it was published on. There is a picker at the top of this page that jumps to any of them. Each update is written from scratch: work that runs for a month stays the headline, but the sentences around it are rewritten every time. We never edit a published edition. When an earlier one got something wrong, the next one says what it corrects.

The October 20 edition printed a right number and drew a wrong conclusion from it, and that is worth saying more plainly than a corrected figure would be. It reported 72 stance positions covering 15 Georgia officials, written at 06:01 that morning, and called eleven days of silence “ended an hour before publication, not yet a restart”. The 72 was exactly right — re-derived this week, that minute holds 72 rows across 15 people. It was not the day’s work. Two more batches followed, at 06:50 and 08:21, and the day closed at 90 positions across 26 Georgia officials and 12 topics. The restart was real; the page measured an hour into it and called it too early.

The same hour caught a second figure mid-flight, and that one is a published count. That edition printed 12,435 curated officials and candidates; it reads 13,184 today. Only 334 of those 749 records were created after it published. The rest were already being written while it counted — 408 people landed in the single hour its query ran, part of a 2026-candidate load that added 823 records across that day. Neither figure was false and both were partial, and the lesson generalises past this page: a count taken while a load is running measures the load, not the work. This edition’s numbers were taken against a quiet database; the last batch written before it — 57 Season 3 answers — finished twenty-one hours earlier.

Four other claims from October 20 were re-derived and every one reproduced exactly. Puerto Rico still has 78 alcaldes with 70 of them lacking a portrait; the trivia bank still reads 3,323 live questions across 43 active collections, split A 25.0%, B 25.9%, C 24.9%, D 24.2%; the certified corpus is still a single stratum of 57 state-level record-based stances; and the unknown-provenance pile is still 245,604 budgets, to the row. That last one reproduced because nothing was done about it.

This pillar is parked and it logged 29 commits in three days, which is a faster rate than last cycle rather than a slower one. Twenty-eight are Civic Spaces and one is a dependency change in Focused Communities; Empowered Listening and Empowered Quests shipped nothing at all. Unlike last cycle, almost none of this is robot maintenance — it is three people building out an Explore view against a Figma file, which is real work on a shell that is already built. It is still not the v4.0 the pivot is waiting for, and it is still happening while the pillar is officially parked. This page keeps counting these commits here rather than in the headline, because a parked pillar with a busy week reads exactly like a reactivated one in a total.

The provenance debt, three days on

A provenance grade is the sentence that says where a budget figure came from and who, if anyone, checked it. Ours run from an audited statement prepared under government accounting standards, through a compilation drawn from audited figures, down to a government’s own unaudited filing — which is what a small town’s budget usually is, and is an honest answer. “Unknown” is not a grade on that scale. It means nobody recorded the answer.

The last edition named this the larger of its two debts and said this page would report it the week it was paid. It has not been paid, and this is that report. Not one of the 245,604 moved. There is no excuse to offer and no progress to describe: the three days went to the stance audit, to campaign-finance transparency in Essentials, and to the broken request. Those were worth doing and this was not done. Saying so plainly is the only thing this page can usefully do about it, and repeating the figure unchanged is how a reader can tell a debt from an intention.

Follow the money

The third leg, alongside votes and stances. The platform holds 29.2 million individual campaign-finance contributions totaling $8.6 billion, covering 2,268 federal politicians and candidates across 24 election cycles (1980 to 2026). Every record traces to its original filing, and superseded lines from amended filings are retired individually rather than counted twice. This cycle the work moved from counting the money to labelling it. A reader can now filter a politician’s contributions to one top donor and open that donor’s own transactions; Los Angeles public matching funds are shown on their own line beside total raised rather than folded into it; loans and one filing schedule are no longer counted as money raised; and the summary footer, the confidence tooltip and the source description now name the actual source instead of implying it is federal. Three of those five are corrections to claims the app was making about its own data, which is the same standard this page holds itself to.

Stance research across the map

Every state has at least one researched official, and all 50 congressional delegations are covered nationally. Shading shows the depth of state and local research. Four tiles moved and one changed tier: Georgia 21 to 32 researched officials, finishing the batch the last edition caught in progress; North Carolina 128 to 132; South Carolina 5 to 9; and Kentucky 4 to 10, which lifts it from Seeded to Growing. Every other tile is where it was. Puerto Rico is still an outlined tile: 161 seats seeded, no stances yet. No tile moved down, which is worth noting in a cycle that withdrew 230 positions — the withdrawals were spread thinly enough that no jurisdiction lost a tier.

Hover any state for its exact count. The seeded tier is mostly governors and statewide officers; deeper tiers add legislatures, counties and cities.

Treasury is still unmeasured rather than low, and the narrowing the last edition spotted did not continue

An earlier edition published a rate here — seven people in a quarter, as 5% of 152 visitors — and withdrew it. It stays withdrawn. The events were sound; the denominator was not, because Treasury pushes its in-app navigation into the address bar without recording a page view, so every visitor is filed against the bare domain. This quarter those few people produced 174 category drills, 104 line-item views, 106 entity selections, 64 year changes, 29 chart changes and 18 searches — up by ones and twos from last cycle, which on a three-day window means almost nothing moved.

102 people reached a budget address and 30 fired any past-the-front-page event. The 102 splits as it always has: 70 landed on our own organisation’s budget, one click from the donate page, which selects nothing and so fires nothing. Asked directly, 36 people opened some other government’s budget; asked by subtraction the same window gives 32, and the gap of four is people who looked at both. Last edition reported this ratio narrowing for the first time and said the day the non-donate half outgrows the rest is the day this section gets a real denominator. It did not narrow. Our own budget was 1.97 times the rest last cycle and is 1.94 times now — flat, not falling. Honouring that promise means printing the number that did not move. It is still not a publishable rate and we are still not publishing one.

The purpose, mapped

An interactive Purpose mind map walks the whole vision — the thesis, and the Why, How, Who and What — with a short explainer on every concept and pillar feature.