Verdict: False
Confidence: High
Last reviewed: October 15, 2026
Original source: Multiple social-media accounts; ultimate origin unclear. The earliest traceable Alberta-cluster instance reviewed here is dated September 14, 2026. No single origin URL is assigned because the card consolidates distinct artifacts.
Short answer
No. The viral artifacts do not prove Alberta's 2026 votes were rigged. The Calgary-Shaw “vote flip” chart was a recoloured Wisconsin 2020 graphic, poll-worker initials are required ballot-authentication marks, and the circulating worker videos came from Canada's 2025 federal election—not Alberta's referendum.
Claim breakdown
Component | Finding | Why |
The viral Calgary-Shaw chart shows Alberta's actual count. | Contradicted | CMTD matched it to Wisconsin's 2020 running-total curve; it had been recoloured in UCP and NDP colours. |
“70 of 71 reporting units” meant 98% of votes had been counted. | Contradicted | Those 70 units contained only 57% of ballots. The final advance-vote unit contained 43%. |
Poll-worker initials can invalidate an Alberta ballot. | Contradicted | Elections Alberta requires the issuing officer's initials and checks them before the ballot is deposited. |
The videos show Elections Alberta workers smuggling referendum ballots. | Contradicted | The footage predates the referendum and comes from 2025 federal-election contexts. One documented clip showed poll-setup materials, not marked ballots. |
Together, these artifacts establish that Alberta's 2026 votes were rigged. | Not established | Altered media, misidentified footage and misunderstood procedures do not demonstrate manipulation of ballots or results. |
What's true
The Calgary-Shaw lead changed when a large advance-vote reporting unit was added. Alberta election officers do initial ballots. Election materials—and, where necessary, sealed ballot boxes—may be transported under prescribed controls. These are real facts, and unfamiliar procedures or delayed reporting can look suspicious without context.
What's wrong or missing
The posts transform those facts into evidence of fraud by altering or relabelling media and by substituting the percentage of reporting units for the percentage of votes counted. They also omit the authentication purpose of ballot initials and the documented jurisdiction and date of the recycled videos.
What the evidence shows
Elections Alberta's official Calgary-Shaw results say 43.44% of ballots were cast in advance and the NDP won by 591 votes.[1] CMTD reconstructed the September 14 reporting sequence: 70 of 71 units represented only 57% of votes; the last advance-vote unit held 43%. It also found that the viral chart was a recoloured Wisconsin 2020 graphic and that repeated versions drew about 260,000 views.[2]
Elections Alberta's 2026 scrutineer guide says voting officers place their initials on issued ballots and election officers check those initials before deposit.[3] Its current counting guidance describes sealed and tracked transport when needed, two-officer counting teams, scrutineer observation, multiple tally checks, and signed and reviewed statements before results are entered.[4]
AFP documented the October 12–13 Alberta posts and traced the marking warning to prior Canadian and U.S. versions. It also established that the worker videos were from 2025 federal-election contexts, not the Alberta referendum.[5] Elections Canada's own safeguards explain federal ballot-box sealing, custody and counting in front of witnesses—the relevant context for the recycled federal footage.[6]
What remains uncertain
The exact activity shown in one Quebec video cannot be independently determined from the clip alone, and public-platform data cannot provide a complete measure of total reach. Alberta's referendum count had not yet occurred when this review was completed, so later authenticated evidence must be assessed on its own. None of those limits turns the present artifacts into evidence of fraud.
What would change our conclusion
We would revise or narrow this conclusion if authenticated Alberta ballot records, chain-of-custody manifests, a recount or audit, a court or regulator finding, or independently verifiable original media showed that ballots were altered, unlawfully transported, excluded or added in a way that could affect an outcome. For the chart specifically, verifiable source data showing it was generated from Alberta's count—not Wisconsin 2020—would overturn that component.
Why it matters to Canadians
The claim family targets trust in an active provincial referendum and can prime audiences to reject results before evidence is available. It also shows how U.S. election-fraud templates migrate into Canadian elections by changing party colours, geography and institutional labels.
Evidence trail
- CMTD / Media Ecosystem Observatory, October 8, 2026: establishes the Calgary-Shaw chart's Wisconsin lineage, the reporting-unit denominator problem, approximate reach and the wider rise in fraud discussion. Its broader platform classification has coverage and model-assisted-method limits.
- Elections Alberta official Calgary-Shaw results, September 24, 2026: verifies the event, winner, margin, turnout and share of advance votes.
- Elections Alberta 2026 scrutineer guide: establishes that official initials are required authentication marks.
- Elections Alberta in-person count guidance: establishes current transport, observation, counting and review safeguards; it describes required procedure rather than auditing every location.
- AFP Fact Check Canada, October 14, 2026: establishes current cross-platform circulation and traces the recycled text and videos. It is corroboration and lineage evidence, not the sole verification backbone.
- Archived primary posts: preserve traceable instances of the poll-worker warning and video relabelling; they establish circulation, not truth, origin or coordination.
- Elections Canada safeguards: establishes the federal procedures relevant to the older footage; Alberta procedure is verified separately.
Review status
Publication status: Ready
Narrative: NAR-20261015-001
Source status: Origin unclear; distinct artifacts have separate traceable instances, but no single origin for the consolidated claim family is established.
Review note: Full-treatment fields and visible page sections were synchronized on October 15, 2026.