AI Riksdag — How should the parties have voted — according to their own documents?

JournalistWhere do parties break from the plan?

Every vote where the AI's document-faithful vote differs from the real one — sorted, searchable, with sources and embeddable graphics.

ResearcherMethods, data and reproducibility

Full methodology, contamination measurement, open prompts, raw data and code.

CuriousSearch a topic you care about

Find the vote, see how your party voted — and what its own plan pointed to.

Trend

Documented divergence from own plan

AllVSMPCLKDMSD

Share of votes per month where the plan's stance differs from the party's actual vote — across all evidence tiers, not only explicit commitments. 0% = the party voted as planned; higher = larger drift. Counts only decisions where the party's plan speaks explicitly to what the vote turned on. The other decisions are still real, fully cited stances — but a plan that never addresses the substance cannot show the party broke it.

Context

Polling support across the term

VSMPCLKDMSD

Monthly average of published opinion polls (SwedishPolls). Context only — not part of the agents' inputs.

Explore

Navigate by time

Each cell is a month: number of votes, colored by AI ↔ actual agreement (green = follows the plan). Click for that month's cases.

By party

Documented divergence per party

V39%83/214 S25%35/139 MP27%60/219 C32%52/163 L83%231/278 KD85%151/178 M68%136/199 SD45%102/227

Share of decisions where the party's plan took an explicit position on the substance and the party still voted the other way. Only the "explicit commitment" evidence tier counts.