Back to browse
SimCongress – a scale model simulation of all 535 congressional members

SimCongress – a scale model simulation of all 535 congressional members

by D_ashe·Mar 3, 2026·2 points·0 comments

AI Analysis

●●SolidBig BrainRabbit HoleNiche Gem

AI Congress simulation exposes why rational bills die—add realistic incentives, bills mysteriously fail.

Strengths
  • Genuinely clever insight: baseline rational agents pass everything, but political self-preservation flips outcomes—validates real legislative friction.
  • Real data integration: VoteView NOMINATE, OpenSecrets, ProPublica, Congress.gov combine into coherent member models with measurable ideology.
  • Full pipeline fidelity: committee markup, rules, floor amendments, filibuster, conference, presidential veto, override—follows actual procedure, not a shortcut.
Weaknesses
  • AI agent modeling details are opaque: how exactly does ideology score translate to vote? Prompt engineering or fine-tuned model? Black box undermines credibility.
  • Crowded space: Niskanen Center, Brookings, and academic papers already simulate legislative outcomes; novelty is the interactive web UI, not the underlying model.
Category
Target Audience

Policy enthusiasts, civics educators, researchers exploring legislative dynamics, people curious about how Congress actually works

Similar To

Niskanen Center's legislative forecasts · FiveThirtyEight's election models · Academic agent-based legislative simulations

Post Description

I built an end-to-end simulation of Congress with all 535 members "modeled".

I’ll share more detail soon, but here’s what surprised me.

In the first version, the agents were simple: voting records, public statements, committee assignments, donor data, and constituent polling. Then I let them vote.

Bills that have stalled for decades sailed through. Getting money out of elections passed with 95% approval. Banning insider trading for members of Congress hit 98%. Ending continuing resolutions cleared 95%.

Turns out, if you model lawmakers as mostly rational actors responding to inputs, they behave… rationally.

To make it realistic, I had to add self-preservation, incentives, and a strong sense of “the other side.”

Now those same bills die in committee. As expected.

Similar Projects

Education●●Solid

AI Guided 3D atomic orbital simulation

Clean 3D quantum viz, but PhET simulations already cover this ground without AI.

SlickCozy
plumocracy
3124d ago
Finance●●Solid

AI Models AI Capex

Stress-test the AI capex bubble with six adjustable levers and five economic shock scenarios.

Big BrainNiche Gem
liujjw
202mo ago