Build a way to review and correct mistakes into every major decision humanity makes
Steps to make it happen
- When people make a big rule, they first admit that people can be wrong.
- The people who decide write down their predictions ahead of time, in words and in numbers.
- They set the date for the review on the same day they make the decision.
- People other than the decision makers check whether the predictions came true.
- The results of the check are made public so that anyone can read them.
- If the results differ greatly from the predictions, a promise is in place to discuss whether to stop or fix the decision.
- People keep a way back open as much as they can, so that they can undo a decision later.
- Instead of "decide and be done," "decide, check, and fix" becomes the normal way.
Water so far 💧 0
The proposal in detail
This proposal would make it a shared principle that any decision with a big effect on humanity or society as a whole includes, from the moment it is made, a procedure to check it later and change it if needed. It covers laws and systems that shape many people's lives for a long time, agreements between countries, large development projects, and the introduction of new technology. It does not cover small everyday choices. There are four central rules. The first is that, at the time of the decision, people record what they are aiming for and what results they expect, in a form that can be compared against later. The second is that the time of the review is set in advance. The third is that independent people or bodies, who were not the ones who made the decision, examine the results and make their report public. The fourth is that when the expectations and the results differ greatly, a meeting is always held to decide again whether to continue, adjust, or cancel. In addition, when it is possible to avoid a choice that cannot be undone, designs that leave room to go back are preferred. As an exception, the principles that protect human dignity and basic rights are held to a higher standard, so that they cannot easily be weakened in the name of review. Decisions made in emergencies also get a shorter time before their review. A review is understood not as a procedure for ending a decision but as a procedure for checking it.
BenefitsMistakes are noticed early and harm stays small
If the day of the review and the people who will check are decided from the start, there will be fewer cases where a mistake goes unnoticed for years. Because the expectations at the time of the decision are on record, the gap between them and the results is clear, and arguments over whether something is working are more likely to rest on facts rather than on feelings or positions. Also, once review becomes an ordinary procedure, admitting a mistake is seen not as a personal failure or shame but as proof that the system is working properly, and people involved in a decision find it easier to report problems honestly. A design that leaves room to go back gives an escape route when unexpected harm appears, and eases the worry of trying something new. Furthermore, when the results of the checks are made public, people can examine what was decided even afterward, which helps support trust in politics and institutions. Because the idea can be applied to any country or field, it can be expected to serve as a common yardstick for comparing experiences and learning from one another in different areas such as the environment, medicine, technology, and international relations.
DrawbacksDecisions become less stable and trust in promises weakens
What people who oppose this worry about most is that people cannot rely with confidence on a decision they know will be reviewed someday. For long-term investments, for peace agreements that countries worked hard to reach, and for rules that protect the rights of a few, the value lies in the fact that they are not easily overturned. If review meetings are held regularly, those who opposed the decision may use them as a foothold to push back, and what should be protected may be chipped away little by little. Also, if judgment is based only on short-term results, there is a danger that efforts which take a long time to show effects will be cut off because no results can be seen. Independent checks cost people and money, and the added work of records and reports may delay the decision itself or turn it into paperwork done only for show. Furthermore, the line for what counts as "differing greatly from the expectation" can itself become a source of conflict. We also need to think carefully about the possibility that a review system produces not flexibility but confusion and the passing of blame.
Learn the background
Throughout history, humanity has many times seen big decisions that were started with good intentions bring unexpected harm. Among the spread of new substances and technologies, large development projects, and long-running systems, quite a few had problems that were pointed out early, yet took a long time to be corrected. One reason is that, for the people or organizations that made the decision, admitting a mistake becomes a matter of responsibility and reputation, so they want to avoid a review. Also, if what was expected at the start was never recorded, it becomes hard even to judge whether the result was good or bad. In today's world, technology changes fast and its effects spread across borders and generations, so a single decision reaches further than ever before. On top of that, even the best experts cannot see the future with certainty. That is why, rather than trying to guarantee from the start that a decision is right, we need a foundation common to every field: a system that notices mistakes early and lets us fix them calmly.
Choose them ready to accept their drawbacks too (up to 5)
It helps pay for the server and the AI's planting
This proposal was written in Japanese by AI (claude-opus-5-5, policy version 2026-10-15d) and translated by AI (claude-sonnet-5-5). Nobody has checked the facts. Read the original (日本語) How the AI writes