Empowered Vote · Volunteer Briefing · September 22, 2026
Every number on this page comes from the live database, verified git history, or our product analytics, not from estimates. We build tools for shared facts; our progress report should be one. From this edition on, each briefing is written fresh and kept at its own dated address, so you can read what we said before and check it against what we say now.
Pennsylvania, South Carolina and Minnesota were seated end to end in eleven days. 823 new officials, every one on a dated term: the Pennsylvania General Assembly at 253 of 253 with 252 portraits, plus Philadelphia, State College, Centre County and twenty county and row offices; the South Carolina General Assembly at 170 of 170 with a full portrait set, plus Columbia, Myrtle Beach and two counties; and the Minnesota Legislature at 201 seats, plus Duluth, Saint Paul and two county boards. Charlotte, Mecklenburg County and Boulder landed alongside them. Each seat was read one at a time against the government’s own roster, which is why these passes keep finding what a bulk import swallows: one portrait had been skipped on a note saying its source site published no usage terms, and the site publishes them, at an address answering the whole time. Nobody had asked it.
Tennessee went from one researched official to 115. A statewide ranked-choice-voting ban seated 74 legislators verbatim, and the Nashville Metro Council was read member by member out of the city’s own legislative record — 40 of 42 seats, both remaining zeros written down as zeros. Maryland moved too: its Cannabis Reform Act seated 116 members on a topic that had been empty. Total sourced stances went 35,877 to 36,420.
And one research pass was found filed under the wrong question. A Transportation Priorities pass on 82 Maine House members had been written into the Residential Zoning slot, which put members who voted to keep parking mandates on screen as broad upzoners. It survived review because one phrase — “reduce parking requirements” — appears on both ladders. The correction is written forward, not erased: the zoning rows are blanked rather than deleted, because deleting them would have let the old answer show through from the previous season. Maine’s stance count rose by 164 and none of it is new coverage. It is the same research, finally filed under the question it answers.
Each update is a dated snapshot. We do not edit one after publishing it.
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. 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 archive starts here. Earlier briefings were published by overwriting this page, so they were not kept. That is the gap this closes, and why the list is short.
Open to everyone, no login. Evidence-based, engaging, anti-partisan.
| Feature | Status | Journey | Where it stands | What's next |
|---|---|---|---|---|
| Empowered Essentials | Live Alpha | alpha |
Where it standsZIP-to-representatives with funding, votes and stances. Eight new communities got a banner. A quieter fix matters more: school-district and county records had been hijacking the Local banner, so a city page could show someone else’s jurisdiction. | What’s nextPortraits for the 70 Puerto Rico alcaldes still without one, and the territory’s first stance research. One sourcing rule was written down: VOTE411 is a pointer, never a cited source. We follow it to the original and cite that. |
| Empowered Compass | Live | alpha |
Where it standsSeason 2 has been what the Compass asks since September 4: 60 topics from a library of 64. A surveillance technology topic was added and used at once to seat eleven Nashville members, and Season 3 exists as a draft — assembled in the open, not switched on underneath the voter. | What’s nextEvery stance row carries a season, a pinned revision and an editor, so a position can change without erasing what was true before. 1,695 answers — about 5% — sit on topics the open season no longer asks: history a voter can read, not a position they can compare today. |
| Treasury Tracker | Live | alpha |
Where it standsThe biggest dataset we hold: 8,144 governments, up 637. The work was identity, not volume. A state publisher renamed a city and we ended up holding it twice, so the loader stopped keying governments on their names: it keys on the publisher’s own stable id and runs a fork check after every load. | What’s nextGrade the rows that still read unknown, and close the per-year document gaps in the loaded roster. The recurring lesson has one shape, and it repeated twice: a check can pass, or fail, and still not be measuring what you think. |
| Read & Rank | In Development | build |
Where it standsA blind taste test for candidates: the question, not the race, is the unit of comparison. No commits for a second cycle. It is gated on the stance pool the research waves are filling, so the three states above are the work that unblocks it. | What’s nextCandidate quote sourcing for 2026 races, once the stance pool for those races is deep enough to draw from. |
| Fallacy Finders | In Development | build |
Where it standsv1.1 is 4 of 6 phases done and built the Fallacy Sleuth, curation-time AI proposing timecoded fallacy tags on a transcript. Its anti-partisan guarantees are mechanical rather than promised: names stripped before the model sees a word, every clip re-run with the speakers swapped and diffed, nothing published without two independent human agreements. Quiet for a third cycle — but the dependency it waits on moved. | What’s nextPhase 10 finishes the US House floor clip end to end, then rights-clean clip sourcing at volume through On the Record. |
| Civic Trivia Championships | Live | alpha |
Where it standsThe Featured shelf shipped, and the label is finally true: the old strip was just the first ten collections in tier order, so the ones it was named for could never reach it. Then the difficulty labels turned out to be decorative. Easy questions are answered correctly 49.9% of the time and medium 50.0% — the same question wearing two labels — while 1,550 medium questions had never been served to anyone, because the default mode draws four of five slots from the easy pool. | What’s nextA rubric now defines the three levels by what the player has to bring, with an easy floor of 25–33%. Relabelling could not reach it, so 59 new easy questions were written — all in the seam every bank was missing, local identity. One collection had sixty-odd questions on council restructuring and the auditor’s office, including its room number, and none asking which river runs through the city. |
Connected accounts, one human one voice. Parked during the election push; resumes after November.
| Feature | Status | Journey | Where it stands | What's next |
|---|---|---|---|---|
| Civic Spaces | Live | alpha |
Where it standsGeo-scoped communities, neighborhood to federal, with friends, posts and role-gated moderation. This was the pillar’s busiest week since April, despite being parked, because the work was interface, not new scope: a three-column shell, and the Issue Alignment Compass brought back drawn with the same component the Compass itself renders rather than a second radar of our own. Members can now read the sibling shards of their own jurisdiction — a city splits into shards past a member cap, and until now there was no way to see what the rest of your city was discussing. | What’s nextTwo decisions, not omissions. A member with no address now gets a real prompt instead of a dismissible banner over a blank page, and it carries no address field, on purpose: Civic Spaces never sees or stores your address. It only reads the jurisdictions the accounts app returns, and that app asks properly. And dark is now the default across every EV subdomain, shared by a cookie on the parent domain, because the old behaviour was a theme that reset every time you switched tools. |
| Empowered Quests | Live | alpha |
Where it standsFact-verification quests with consensus scoring; verified facts already feed Essentials. Red gem quests unlock at VR 90+. | What’s nextv1.6 planning. |
| Focused Communities | Paused | build |
Where it standsv1.0 shipped April 17: per-topic forums wired to the Compass. Waiting on the post-election Connect pivot. | What’s nextv1.1: badge ratification and argument mapping. |
| Empowered Listening | Paused | build |
Where it standsStructured debate infrastructure; 5 of 6 phases done — timers, transcription, audience notes. Its planning documents were brought back in step with what is actually built — the cheapest thing you can do for a project you mean to pick up again. | What’s nextPhase 6: voting and badges. Then Common Grounds can build on it. |
| Emparks | Design | design |
Where it standsPlatform-wide annotation system, five mark types. Full design doc; no build yet. | What’s nextAwaits Connect-pillar capacity. |
| Common Grounds | Design | design |
Where it standsAdversarial-collaboration mediation. Full design doc; depends on Empowered Listening and Fallacy Finders arbiters. | What’s nextBuilds after its dependencies land. |
The shared spine every app plugs into.
Auth, tiers, XP and gems, politician database, TIGER geofencing, elections. This cycle it wrote its privacy rules down instead of holding them in people’s heads. A decision record sets the privacy floors — what the platform may show about a person, and when — with a retention policy for the stance audit, so a read of someone’s answers is bounded by a rule anybody can look up rather than by whoever wrote the endpoint. The audit’s pre-flight checks ran before it was built and found the first stage of the design wrong, which is what a pre-flight is for. Maryland’s state-legislative district identifiers had collapsed onto each other — 84 rows, not the 24 first reported — and the repair carries a guard so the shape cannot regress quietly. 9,189 districts mapped, 121,178 legislative votes and 19,622 bills on record.
Speaker-attributed, searchable transcripts of public meetings, synced to the original video. It restarted. After five cycles of near-silence and two consecutive unexplained drops in the segment count, ten meetings landed — legislative days, a city council regular session, a full interview, and an LA mayoral debate — and the dataset now reads 164 meetings, 18,923 transcript segments and 1,190 identified speakers, including LA City Council with 7,200 recorded council votes. Two things to be straight about. The segment total is now above where it stood before the drops, so the new ingest more than covered them — but the drops themselves are still unexplained, and a total moving up does not answer why it moved down. And segments carry no creation timestamp, so the ten dated meetings are the evidence here, not the segment count. Built by volunteer Chris Andrews; it feeds the meetings system and the clip pipeline Fallacy Finders is waiting on.
Treasury Tracker holds a set of frozen figures: budget numbers that have been verified and may not move without a recorded decision. A weekly invariant compares the live database against that set. It failed on September 7, 14 and 21 — three weeks running — and nobody was told. The job fired perfectly every time and wrote each verdict into a table nothing in the codebase reads, and the workflow that would have raised an issue had been switched off for want of a credential. The check worked. It reported into a void.
The data was never wrong. The database held an expectation three revisions out of date while the repository had moved on through decisions recorded properly each time; the differences match those notes exactly, which makes the diagnosis certain rather than plausible. The real residual was two rows, not sixty-five. Three things changed: the verdict is now read on every daily run, and a missing or stale one counts as a failure rather than a pass; the runner no longer reports one condition when it has found another; and a live privilege sweep now runs inside the database weekly, because the old guard read the text of code changes and could never see anything made through a console. We would rather publish this than a clean week. A monitor nobody reads is worse than no monitor, because it costs you the belief that you are being watched.
The invisible work. This is what "evidence-based" costs, and we paid it.
36,420 stance positions across 4,196 politicians and 64 topics, every one carrying reasoning and cited sources; 1,690 of the 2026 candidates already have some. One figure needs its date said out loud: 2,722 sources were independently re-checked in a sweep that finished in April 2026, and the number has not moved since, because every row in it was written then. It is a dated achievement, not a running total.
| Coverage area | Politicians | Stances |
|---|---|---|
| US Congress + national | 571 | 9,647 |
| 2026 candidates & discovered names (no seat) | 940 | 6,044 |
| California (state + local) | 373 | 4,787 |
| Massachusetts (state + local) | 302 | 3,355 |
| Maryland (state + local) | 188 | 2,228 |
| Texas (state + local) | 270 | 2,073 |
| Maine (state + local) | 186 | 1,794 |
| Virginia (state + local) | 166 | 900 |
| Oregon (state + local + county) | 154 | 824 |
| Wisconsin (state + county + local) | 208 | 628 |
| Utah (state + city + county) | 48 | 555 |
| Washington (state + county + city) | 161 | 489 |
| Arizona (state + local) | 129 | 436 |
| North Carolina (state + local) | 124 | 406 |
| 39 more jurisdictions (incl. Tennessee, Florida and DC) | 376 | 2,254 |
The table reconciles, and that is a weaker check than it looks. Every politician with at least one answer lands in exactly one row, and the rows sum exactly to the headline totals. That catches a missing row. It cannot catch a row filed in the wrong place, because a misattribution moves a count between buckets without changing the sum — which is how a past error survived a table that reconciled perfectly. The corroborating signal is agreement across cycles: rebuilt today, thirteen of the fourteen named rows reproduced their politician counts exactly. Tennessee is why the remainder row grew by 118 people, and it has no row of its own yet; next cycle it should.
How a stance gets seated. A member is placed on a topic because of a bill they sponsored, read in its session-law text — not a sentence someone else wrote about them. Three rules travel with every wave: a unanimous bill cannot evidence a position; cite the session law, not the introduced bill, because a bill can be gutted by a committee substitute long after members file it and a sponsor list never changes; and where a mechanism matches no rung clause by clause, the answer is blank. Two blanks this cycle are worth more than most of the seatings: one Nashville member’s record was read in full and yielded zero, and Tennessee’s eight education topics were triaged as unreachable from state statute.
Treasury Tracker’s database holds budget data for 8,144 government entities across all 50 states plus DC, spanning fiscal years 1976 to 2028:
Empowered Vote’s budget is published in Treasury Tracker like any other entity, and putting ourselves through our own product found it describing us wrongly. The line under every hero headline read “Explore how public funds are allocated and spent” — including on our page, where the money is donations, not public funds. No figure was wrong, which is why nothing would have flagged it. The page now says what it is, and deliberately makes no completeness claim: it publishes the bank balance, and we also hold an untransferred wallet balance, so “every dollar” would be an overstatement. A transparency page is the worst place to overstate coverage.
The third leg, alongside votes and stances. The platform holds 28.2 million individual campaign-finance contributions totaling $7.8 billion, covering 1,302 federal politicians and candidates across 24 election cycles (1980 to 2026). Every record traces to its original FEC filing. The pipeline handles amended filings — superseded lines are retired individually instead of counted twice. This is the “who funds whom” data behind Essentials, presented without commentary.
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. Tennessee’s 1 to 115 is the largest single-cycle move this map has recorded, and Maryland deepened to 188. Minnesota, Pennsylvania and South Carolina barely moved, which understates them badly: all three had hundreds of seats loaded, and this map shades by stances researched, not seats held. Puerto Rico is an outlined tile — 161 seats seeded, no stances yet.
Hover any state for its exact count. The seeded tier is mostly governors and statewide officers; deeper tiers add legislatures, counties and cities.
Who actually uses what we build — measured, not guessed.
609 people used the platform in the last 90 days, across every app, over 1,703 sessions. What they do once they arrive is good — sessions average six and a half minutes and 7 in 10 don’t bounce — so the constraint is arrival, not engagement. 93 of the 300 who reached the main site clicked through into an app, 31% in the quarter.
Read every figure here as an order of magnitude, not a trend line. At this volume a couple of people either side is the entire swing, and the quarter window slides forward, so an app can “fall” because a busy week in June rolled out of it. Nothing this cycle was aimed at arrival, so 635 to 609 is the expected result of not working on it. The hostname map was re-run before any of these figures were pulled and found no new host, which is what makes this pull readable against the last four: the main site went 306, 316, 312, 297, 300. One host that served nobody last quarter served somebody this one — the argument for leaving quiet hosts in the filter rather than tidying them out.
Everyone who arrived, against the people who did the one thing that app exists for. Bars are the 90-day rate; counts are people.
Two of these are start problems, not finish problems. Only 17 people started a trivia game in a quarter, on 110 arrivals; 12 finished. Read & Rank is the same shape: 23 races started, 10 ballots revealed. The Compass is the opposite and is the healthiest funnel we have: 114 of its 126 visitors opened the quiz and 35 saw it through.
Unique visitors per app, 30 days and the full quarter. Each app is counted once across every hostname that serves it — Treasury draws traffic on four, the main site on two.
Treasury Tracker is still the second most-visited thing we run and holds the platform’s largest dataset. Civic Spaces is absent from this table, and that is a measurement gap rather than a result: it serves no traffic our analytics can see. Whether Treasury’s 145 reached a budget is the question the note below is about.
An earlier edition said seven people in a quarter opened any of Treasury Tracker, as 5% of 152 visitors. That rate was not sound and it stays withdrawn. The events were fine; the denominator was not — Treasury pushed its in-app navigation into the address bar without recording a page view, so every visitor was filed against the bare domain. The direction was wrong too: this quarter those few produced 110 category drills, 82 entity selections and 61 line-item views. The handful who reach a budget use it hard.
The fix has been recording since August 16, and the two ways of counting still disagree by five times. 52 people reached a budget address; only 11 fired any past-the-front-page event. The split explains it: 41 of the 52 landed on our own budget, one click from the donate page — a transparency visit, not budget exploration, and it selects nothing, so it fires nothing. That leaves 12 who opened some other government’s budget, down from 16. Neither is published as a rate. Treasury stays unmeasured rather than low until the half that is not ours outgrows the donate traffic. The rule it bought applies to every app here: a low count is a claim about instrumentation until you have read the code path.
How much further we have to go, honestly.
Research the seats just loaded — Pennsylvania, South Carolina and Minnesota have rosters and almost no stances. Grade the budget rows still marked unknown. Finish the 2026 House candidate waves and land Read & Rank candidate quotes once the pool can carry them. Success in November looks like one thing: any voter in a served area can look up every name on their ballot and follow the money behind it.
Reactivate the parked pillar: Empowered Listening phase 6, Focused Communities v1.1, Civic Spaces v4.0 — whose shell, shared theme and sibling-shard view are groundwork done early. This is where the platform stops being a reference tool and becomes a community, and where our sustaining-donor model lives.
Designed but not built: the Gems stipend economy, Awareness Exchange, Badges, Symposium tournaments, Emparks, Common Grounds, and tools for citizens to run for office on the strength of their solutions. Roughly a third of the founding vision, waiting on the first two pillars.
A volunteer-built platform, visible in the commit history.
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.