GroundedVote
Find out who
actually matches
your values.
Not your party. Not your tribe. Not your fear. Your actual policy values.
GroundedVote is a nonpartisan civic alignment engine. It uses a bias-audited AI pipeline to generate neutral policy questions from publicly available candidate data — then shows you which candidates most closely match what you actually believe.
The Problem
Voter disengagement is not a motivation problem. It's an information architecture problem.
Identity Signaling
Voting has become a performance of group membership. Americans increasingly vote against outgroups rather than for policies.
Fear Activation
Threat-based messaging suppresses deliberative processing. Voters in high-fear states are significantly less likely to weigh policy tradeoffs.
Party Loyalty
Party affiliation functions as a heuristic that bypasses policy evaluation entirely. The party label is a sufficient signal — for most voters.
Media Distortion
The issues that dominate news coverage are not the issues that most affect most people's daily lives. Culture war topics generate clicks at rates disproportionate to their material impact.
“The problem is not that voters are irrational. The problem is that the information environment systematically rewards tribal activation over policy alignment.”
GroundedVote Founding Framework
Who This Harms Most
Not low-information voters by choice. Low-information voters by design.
The voters most harmed by the current information architecture are those with the least institutional access: first-generation voters, low-income voters, voters in communities with weak local journalism, and those who have experienced repeated civic disappointment.
GroundedVote is built specifically for this population. Accessible to someone who has never followed an election closely. Engaging enough to complete. Trustworthy enough to share.
First-generation voters
Low-income voters
Communities with weak local journalism
Voters who have experienced repeated civic disappointment
HBCU students and communities
Young voters entering civic participation for the first time
The Method
Three phases. One civic mirror.
Phase 01
Data Collection
Candidate position data collected from official government records, campaign platforms, verified media interviews, and third-party aggregators. Source hierarchy strictly enforced.
Phase 02
Bias-Audited Questions
A three-pass multi-model AI pipeline generates and audits neutral policy questions. Loaded language, false equivalence, and asymmetric framing are scored and eliminated before any question reaches a user.
Phase 03
Civic Alignment
You answer a weighted issue quiz calibrated to real population impact. The result: a civic mirror showing which candidates most closely match your actual beliefs — with full methodology transparency.
The Civic Mirror
The quiz is coming.
GroundedVote is in development. The full platform launches ahead of the 2026 election cycle. Join the list to be notified when it goes live.
Civic partners, funders, and election officials interested in the methodology can request the whitepaper below.
No spam. Notification only. Unsubscribe anytime.
For Partners & Reviewers
Read the Methodology White Paper
Version 1.0 · April 2026 · RhetoricalPoints LLC · Minneapolis, MN