There are no easy answers here. The proposal should earn trust through hard questions, careful tests, and changes made in public.
A proposal to test, not rules already in force. No constitution has been adopted. There are no verified constituents, binding elections, or common treasury operating through this site. AI would help people do the work; people would hold the authority.
These are research plans, not adopted rules.
No verification method, voting threshold, legal structure, or treasury has been selected or activated. Confirming an email would not establish unique personhood or create voting rights.
Each question below names a way to investigate it. A result that challenges the proposal should be published alongside one that supports it. Some findings may show that a proposed mechanism should not proceed.
People, privacy, and a fair say
How do we count each person once without watching them?
Proposed research approach
Proposal v0.1
Compare personhood-check methods in a small, voluntary research trial. Test duplicates, false rejections, data retention, and appeals before selecting any method.
Purpose
Equal civic standing needs a fair way to prevent duplicate participation while protecting privacy.
Tradeoff
Less personal data can make duplicates harder to detect. More proof can expose people or exclude those without documents.
Open question
Can any approach meet privacy, equal access, and one-person requirements well enough for binding use?
Who should have civic standing, and when?
Proposed research approach
Proposal v0.1
Publish competing eligibility models, including age, residence, affected interests, and entry or exit rules. Test them with people likely to be included and excluded.
Purpose
A group should explain who its decisions affect and why those people can or cannot vote.
Tradeoff
Broad eligibility may weaken local voice. Narrow rules can shut out people who bear real costs.
Open question
What boundary is fair for a voluntary institution working across places with different laws?
How can people take part when daily life gets in the way?
Proposed research approach
Proposal v0.1
Test short materials, translation, accessible formats, flexible schedules, and support for panel service. Ask nonparticipants what stopped them.
Purpose
An invitation alone does not give people time, safety, internet access, or confidence to speak.
Tradeoff
Support costs money. Gathering information about barriers can itself expose private circumstances.
Open question
Which supports change who has a real say, rather than only increasing attendance?
Rights, power, and trust
Which rights are protected, and who settles disputes?
Proposed research approach
Proposal v0.1
Draft a rights floor and competing review models with legal specialists and people who have faced exclusion. Test hard cases and publish minority views.
Purpose
A majority must not be able to remove basic protections through an ordinary policy vote.
Tradeoff
Reviewers need independence, but they must not gain unlimited power to decide policy.
Open question
Which rights may never be removed, and how should a mistaken rights ruling be corrected?
How hard should changing the constitution be?
Proposed research approach
Proposal v0.1
Compare supermajorities, waiting periods, turnout rules, and approval across groups or places. Use worked examples with different sizes and participation levels.
Purpose
Rules must be able to improve without letting one organized bloc rewrite them for itself.
Tradeoff
A high threshold can protect rights but also lock in a harmful rule. Approval across groups can create small-group vetoes.
Open question
Which thresholds protect against capture while still allowing a fair, needed change?
What makes the first institutions legitimate?
Proposed research approach
Proposal v0.1
Publish a temporary founding mandate, independent transition review, proposed ratification stages, and conditions for stopping. Test representation rules against unequal participation.
Purpose
The founder's approval and early supporters' interest are not consent from a verified constituent body.
Tradeoff
A careful start is slow. Early participants may still shape the rules before excluded people can join.
Open question
Who can fairly authorize the first vote, and what prevents the earliest or largest bloc from dominating it?
Law, money, and responsibility
What legal form could support this work?
Proposed research approach
Proposal v0.1
Commission jurisdiction-specific advice on legal entities, privacy, contracts, governance duties, and dispute resolution before any activity begins there.
Purpose
An internal constitution does not replace local law or create government powers.
Tradeoff
Different places may require different entities or restrict participation. Advice can identify limits that cannot be designed away.
Open question
Which activities are lawful where, who is accountable, and which court or appeal body can offer a remedy?
How can a common treasury be funded without selling control?
Proposed research approach
Proposal v0.1
Compare funding limits, donor disclosures, reserves, independent custody, conflict checks, and staged approvals with qualified financial and legal reviewers.
Purpose
Resources should support the mission without giving funders hidden decision rights or putting money beyond lawful oversight.
Tradeoff
Strict limits may reduce funding. Public financial reporting must still protect private and legally restricted information.
Open question
What funding and custody rules can survive dependence on large gifts, financial losses, or a disputed grant?
Who bears the cost when a lawful decision causes harm?
Proposed research approach
Proposal v0.1
Test stop rules, complaints handling, whistleblower protection, independent audit, and remedies using failed-project scenarios.
Purpose
Following procedure does not remove responsibility for a poor result.
Tradeoff
Promises of compensation can create costs the group cannot meet. Some harms cannot be reversed.
Open question
Who has authority to pause work, what remedy is owed, and who pays for it?
Voting, evidence, and AI
What voting technology is safe enough for each decision?
Proposed research approach
Proposal v0.1
Begin with nonbinding, low-stakes research. Require independent security, accessibility, legal, and coercion-resistance review before any binding control. Compare offline and hybrid options too.
Purpose
People need private choices, correct tallies, fair access, and a way to check and challenge results.
Tradeoff
Safer methods may be slower or harder to reach across borders. A receipt that proves a choice could also enable coercion.
Open question
Can a future method let people check inclusion without revealing choices, and what failures must stop certification?
How can evidence and AI work be checked without shared blind spots?
Proposed research approach
Proposal v0.1
Test sourced summaries against original material, competing arguments, independent human review, and models with different design choices. Publish errors and corrections.
Purpose
Agreement between AI systems is not independent proof when they may use the same sources or assumptions.
Tradeoff
Human review and truly different evidence cost time. More models can add noise rather than reliable checks.
Open question
How do we know an important objection has survived summarizing and translation, and when must a decision be reopened?
Where must the line around AI authority be drawn?
Proposed research approach
Proposal v0.1
Test explicit task limits: source-linked explanation, translation, comparison, and audit assistance. Keep consent, ballots, final certification, and spending authority with accountable people.
Purpose
A useful tool must not quietly become the actor that decides what people want or what rules mean.
Tradeoff
Human approval can become a rubber stamp. Strict limits can reduce speed and still leave subtle influence unchecked.
Open question
What decisions should be off limits to AI assistance altogether, and how can hidden steering be detected?
Help us find the weak point
You do not need to endorse this proposal to review it. Practical experience, a clear objection, and a better alternative all matter.
Choose a specific rule, example, or missing protection.
Describe who could be hurt or left out, and how.
Point to evidence or a case that tests the claim.
Suggest a change, name its cost, and say how we could test it.
Current status: this site does not collect critiques or signups. A future critique channel is not open yet. You can read and prepare a review of this draft now; the homepage shows participation status.