The strongest objections belong in the proposal, not in the fine print. These risks remain even if everyone starts with good intentions.
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.
Proposed safeguards, not current operating systems.
No governance bodies, election system, or common treasury described here are active. A safeguard on this page is a design to test. It is not a security assurance.
Open any risk to see the proposed response, its costs, and the part it does not solve. Ask whether the check still works when the people in charge want to evade it.
Who could take control?
1. The founder keeps control
Proposed response
Proposal v0.1
Give the founder no permanent veto. Limit terms, publish conflicts, and put removal and appeals under independent bodies.
Purpose
A project can promise shared power while still depending on one person's money, name, or approval.
Tradeoff
A handover may slow the work or lose useful knowledge. Independence needs its own funding and staff.
Open question
Who can oversee the first handover before independent institutions exist?
Residual risk: A founder could still control hiring, funding, software access, or public attention without holding formal office.
2. Donors buy influence
Proposed response
Proposal v0.1
Give donations no extra votes or seats. Publish funding and conditions. Review conflicts and refuse gifts that buy control.
Purpose
Large gifts can steer the agenda even when voting rights are equal.
Tradeoff
Turning down money may limit what the group can do. Small gifts also cost time to manage.
Open question
What funding limits prevent dependence on one donor without making the project unworkable?
Residual risk: Donors could fund outside campaigns, favored experts, or related organizations to gain indirect influence.
3. Experts become the rulers
Proposed response
Proposal v0.1
Require competing evidence, published conflicts, citizen-panel questions, and public reasons. Experts advise; accountable people decide.
Purpose
Technical knowledge should help people make choices, not take those choices away.
Tradeoff
Competing reviews take time. People may struggle to judge a dispute between specialists.
Open question
How can a panel test expert claims without just trusting a second set of experts?
Residual risk: A small professional network could share the same assumptions, control the language, and narrow the choices.
4. Officials and contractors make a deal
Proposed response
Proposal v0.1
Separate approvals, custody, audit, and oversight. Publish contracts and conflicts. Set reporting and removal duties.
Purpose
People who approve work should not secretly share in the money or shield each other from review.
Tradeoff
More checks mean more staffing and slower purchasing. Small teams may find true separation hard.
Open question
Who investigates when several oversight bodies are involved in the same deal?
Residual risk: People across separate bodies could still collude, hide favors, or retaliate against someone who reports them.
Whose voice could be lost?
5. One person appears to be many
Proposed response
Proposal v0.1
Test ways to verify one eligible human once, keep identity checks separate from ballots, and provide independent appeals.
Purpose
Fake or duplicate identities could turn equal voting into a numbers game. These are sometimes called Sybil attacks.
Tradeoff
Stronger checks can invade privacy or exclude people who lack documents. Email confirmation does not solve this.
Open question
What proof is enough without creating a store of sensitive identity data?
Residual risk: Identity systems can make mistakes. Real people can also sell access or face pressure to act for someone else.
6. Only people with spare time take part
Proposed response
Proposal v0.1
Test translated materials, flexible schedules, accessible formats, participation support, and fairly sampled citizen panels.
Purpose
An open invitation is not enough when time, care work, disability, cost, or internet access keeps people out.
Tradeoff
Support costs money. Panel selection still involves hard choices about which differences to reflect.
Open question
How do we measure who is missing and change the process without collecting too much personal data?
Residual risk: A process can meet its attendance targets while people still feel unable to speak or influence the result.
7. A loud campaign overrides rights
Proposed response
Proposal v0.1
Keep protected rights above ordinary majorities. Use waiting periods, public evidence, minority reports, and independent rights review.
Purpose
A popular campaign or simple slogan can push a group to punish opponents or skip basic safeguards.
Tradeoff
Waiting and review can frustrate people who want urgent change. Reviewers can also misuse their role.
Open question
What can a rights reviewer block, and how can that ruling be fairly challenged?
Residual risk: A majority could try to replace reviewers, reshape eligibility, or amend the rules to silence a minority.
8. Fair rules hide unfair treatment
Proposed response
Proposal v0.1
Pair formal audits with interviews, access testing, complaints review, and minority reports about people's real experience.
Purpose
A legally valid vote does not prove that everyone could take part safely or that the outcome was fair.
Tradeoff
Experience is hard to measure. Reports can conflict, and protecting privacy can limit public detail.
Open question
What evidence should trigger a new review even after every formal rule was followed?
Residual risk: Officials could celebrate good statistics while overlooking intimidation, unequal burdens, or quiet exclusion.
How could information be bent?
9. AI persuades instead of informs
Proposed response
Proposal v0.1
Prohibit targeted AI pressure and inferred consent. Label AI work, show sources, and give people access to opposing arguments.
Purpose
A helpful-looking assistant could quietly steer choices using what it knows about each person.
Tradeoff
It can be hard to distinguish helpful explanation from persuasion. Enforcement needs independent tests.
Open question
Who can test for covert steering without exposing private conversations?
Residual risk: A system could shape choices through tone, order, repetition, or omitted options even without an explicit instruction.
10. A summary leaves out the strongest objection
Proposed response
Proposal v0.1
Link summaries to original material, preserve dissent, publish corrections, and commission human review of contested summaries.
Purpose
People may rely on a short version that makes one side look stronger than the evidence supports.
Tradeoff
Reading original material takes time. More reviewers do not guarantee different points of view.
Open question
When does a missing argument require a new decision rather than a correction note?
Residual risk: Several models may repeat the same blind spot. Most readers may never check the source.
11. False evidence gets into the process
Proposed response
Proposal v0.1
Track sources and changes, require conflict disclosures, test disputed claims, and allow independent challenges before decisions.
Purpose
Forged studies, altered records, or hidden instructions could poison the evidence used by people and AI.
Tradeoff
Checking sources costs time and can unfairly favor established voices over new findings.
Open question
What is enough evidence to pause a decision when a key source is disputed?
Residual risk: A convincing false claim may survive review. Genuine evidence may also be dismissed too quickly.
12. An attack changes records or stops participation
Proposed response
Proposal v0.1
Before binding use, require independent security tests, controlled access, recovery drills, tamper checks, and audited fallback procedures.
Purpose
An attacker could alter eligibility, steal private information, change tallies, or stop people from taking part.
Tradeoff
Stronger security costs money and may reduce ease of use. A fallback can itself exclude people.
Open question
What evidence would justify using any online system for high-stakes binding votes?
Residual risk: Compromised devices, insiders, and coercion remain threats. Secure, globally accessible, coercion-resistant Internet voting is not a solved problem.
What could go wrong after a decision?
13. Administrators hide what they are doing
Proposed response
Proposal v0.1
Publish mandates, reasons, contracts, spending, and results. Provide independent audit, complaints review, and protected reporting channels.
Purpose
People cannot challenge a decision they cannot see or understand.
Tradeoff
Full disclosure can expose personal data or sensitive work. Records also take effort to keep useful.
Open question
Who reviews a claim of confidentiality, and what minimum record must stay public?
Residual risk: Officials can bury key facts in long reports, delay publication, or make an appeal too costly to use.
14. Temporary emergency powers become permanent
Proposed response
Proposal v0.1
Set narrow powers, spending caps, prompt review, and automatic expiry. Require a fresh lawful decision for any renewal.
Purpose
An emergency should not become a standing excuse to bypass the rules.
Tradeoff
Strict limits can slow a real response. Repeated review needs people who remain independent under pressure.
Open question
Who can stop emergency action immediately when ordinary reviewers cannot safely meet?
Residual risk: Leaders could repeatedly declare a crisis or renew powers until the exception becomes normal.
15. The proposal conflicts with local law
Proposed response
Proposal v0.1
Require legal review in each place where activity is proposed. Act only through accountable humans and lawful entities; stop actions that lack lawful authority.
Purpose
An internal vote cannot grant powers that an organization does not legally hold.
Tradeoff
Different laws can limit shared plans. Some proposed activities may not be feasible at all.
Open question
How can people appeal an internal decision when more than one legal system applies?
Residual risk: Laws can conflict or change. An internal process may offer a remedy that a local court will not recognize.
16. A proper process still makes a bad choice
Proposed response
Proposal v0.1
Use small pilots, staged spending, outcome checks, clear stop conditions, and routes to reconsider or reverse decisions.
Purpose
Good procedure cannot promise that a policy, grant, or project will work.
Tradeoff
Pilots can delay useful work. Some losses or harms cannot be undone.
Open question
Who can stop a failing project, and what remedy is owed to people harmed by it?
Residual risk: A poor outcome can arrive before review catches up. Those with power may defend the decision instead of learning from it.
Try the hardest case
Choose a risk. Write down who benefits, which check they would try to evade, how someone could detect it, and what should happen next. Include the strongest reason your proposed fix could fail.