Polis of the discussion in Mr. Hoshi's PR
Polis of the discussion in Mr. Hoshi's PR
Discussion on "Creation of a new section on the advancement of the human rights situation (proposal by Akio Hoshi)
30 comments that ChatGPT o3 can vote for or against based on the overall discussion on the new item on advancing the human rights situation (proposed by Akio Hoshi) received in Team Mirai's manifesto.
2025-08-13
nishio The original assertion (0) is highly reserved, and is an area of uncertainty, and the term "human rights" is highly abstract and is easily interpreted differently by different people, causing sterile discussions. The term "human rights" is highly abstract and is easily misinterpreted by different people, leading to sterile discussions. I wonder if we should ask AI to make a reworked agenda and create a new Polis. https://pbs.twimg.com/media/GyMkvN3aQAAinHt?format=jpg&name=large#.png https://pbs.twimg.com/media/GyMlXGIaEAQUfkN?format=jpg&name=large#.png
nishio Also, most people agree that the concept of human rights should not be denied, and most people also agree that we should discuss individual issues rather than lumping them under "human rights" and debating whether to restrict or promote them. Most people agree that we should not deny the concept of human rights. https://pbs.twimg.com/media/GyMpWVXaEAQnN08?format=jpg&name=medium#.png
I spoke to Mr. Hoshi via DM.
I told him that I was interested in "the mechanism of deliberation" and not in "human rights," and that when I made the next Polis, I felt like making a Polis that said, "It was ill-advised to put human rights on the agenda, let's limit the discussion to selective surname selection.
https://gyazo.com/4d9d272136df92028aa6c97fee644406https://gyazo.com/bea6ce31d3221537e3f64192bd2c15a5
2025-08-10
2025-08-09
2025-08-08
https://gyazo.com/2a40c70f8dbd75ddfeba0c6effbe9d4a
nishio.iconThis is a Polis report. Suggest some COMMENTS that might be good to add to bring the fragmentation closer to resolution, with reasons.
GPT5.iconCopy. Looking at the report, the axes on which agreement is likely to be gathered are "clarity of definition, emphasis on evidence, and transparency of procedures" (#13 #29 #34 #60 #61 #54, etc.) and the divisions are "family surnames, treatment of international standards and means (such as referendums)" (#1 #7 #14 #27 #28 #26, etc.).
I will provide a "proposed additional comment" with reasons (short text that can be pasted directly into Polis) to bridge the gap.
1. "Advancing the Human Rights Situation" will first establish agreed-upon evaluation indicators (e.g., reduction of discrimination victimization, child welfare, reduction of procedural burdens) first.
Rationale: Avoid value clashes and rectify discussions based on outcome criteria (works well with #13). 2. specify at the outset the working definitions of "tradition," "culture," "public welfare," and "human rights" to be used in this discussion.
Rationale: Reduce misplaced words and prevent sterilization (reinforces #29 #43).
3. cited studies should have COI disclosure and source credibility criteria, and primary sources for both approval and disapproval should be listed together.
Rationale: Preemptively mitigate distrust over sources of information (consistent with majority agreement with #62 #59).
4. family and family name system proposals will be pre-evaluated using "best interest of the child" indicators (bullying, sense of belonging, administrative burden, etc.).
Reason: A's emphasis on "security of family and children" and B's "self-determination" can be compared with the same yardstick (bridge between #49/#52 and #14).
5. institutional changes piloted in limited areas/periods → independent evaluation → national rollout/modification, with automatic review (sunset) every 3 years.
Rationale: Allow room for progress while taking into account concerns about poor performance (materialize the area of uncertainty in #24). 6. when introducing a choice system, the UI, cost, and effort of choosing same/separate surname should be equivalent, and the design guidelines should be clearly stated to avoid implicit pressure.
Rationale: Simultaneously ensures B's pressure concerns (#15) and A's respect for free choice.
7. international human rights standards are "principle-referenced + if there are differences, the reasons are made public. When there is a deviation, accountability is imposed on the parliament.
Reason: Mediate the conflict in #28 with "procedures" (both sides are in the same face). 8. rephrase the discussion to specific examples and indicators, rather than using labels such as "discrimination" and "human rightsism" in the conclusion.
Rationale: Simultaneous treatment of #9 agreement and #38 splinter source, reducing mutual attacks. 9. make the constitutionality checklist of the party and the Legislative Bureau publicly available for citizens to follow up.
Rationale: "Visualize" the reliability of the existing process (bundle #12 #31 #36 #34). 10. for AI use of policy documents, the model name, input text, and output logs should be disclosed, and it should be clearly stated that the final responsibility lies with humans.
Rationale: Both the axis in favor of #11 and the risk perception of #25. 11. ensure equal hearing slots for both pro and con parties (religious and traditional communities/rights parties)
Rationale: To implement #54 majority consensus and reduce distrust of representativeness. 12. make minority impact assessments and comparisons of alternatives mandatory for matters involving fundamental human rights before they are subject to referendum
Rationale: #26 conflict is dealt with by agreeing on "applicable criteria" and clarifying minority protections. I've posted it as a reference and put it in my own words.
2025-08-07
nishio @AkioHoshi I made Polis I will take a closer look and respond.
I think you extracted this from a discussion on my PR, but the question itself contains too many layman's and polemical arguments, which may cause people to be concerned.
I recommend that the questions be scrutinized by a legal professional who understands human rights issues. If not, then make it private first.
AkioHoshi I say this out of real kindness, but it is not good for the reputation of "Team Mirai" to publish this now. (If you're not sure, why don't you listen to the PR proposer first? I'm willing to help if you need professional connections)
dmikurube These appear to be "questions", as they present the "opinions of existing respondents" to the current respondents and ask for their approval or disapproval, and then use their approval or disapproval to map the opinions of the current respondents. This "question" does not seem to have been prepared by the party in question, just in case. Although it may be true that the "opinions of respondents who have already responded" are a good indication of the propensity of the respondents to support the survey.
nishio I consulted with a legal professional who understands human rights issues regarding Mr. Hoshi's concerns and added a note of caution. (As Mr. Sanzabe pointed out, we took the form of stating that so as not to cause misunderstanding.) As I stated, anyone can add their opinions, so if there is a lack of arguments, Hoshi can add his/her own.
If you would like to create the initial questions by experts instead of doing so, that would be interesting as well. We would be happy to work with you towards creating the "next version" of the questions created by the experts.
I will not keep this version private as it is still yielding interesting findings at this time. I will do a separate post explaining the interesting points after this.
explanation
nishio Regarding Hoshi's proposal for Team Mirai's manifesto, which got the comment section very excited, I extracted opinions from the comments posted using ChatGPT o3, It is very interesting that there are three opinion groups. https://pbs.twimg.com/media/Gxt1ZrIboAAeUAg?format=jpg&name=medium#.png
nishio Here's what I have to say about the common opinion. https://pbs.twimg.com/media/Gxt1jM0awAMdgSs?format=jpg&name=medium#.png
nishio Here is what I had o3 explain about the 3 groups It can be read that the participants are divided into three extremes: the free speech camp (A), the human rights proponents (B), and the cautious conservatives (C), each participating in the discussion under different value systems.
A: "Labeling NG" and "AI must be transparent" - emphasis on creating an environment for free discussion.
B: "Actively promote education on married couples and human rights" - liberal and pragmatic (competitive).
C: "Concerned about excessive human rightsism" and "cautious about changing the system" - conservative and sovereignty oriented.
nishio It is not a dichotomy between "human rights proponents" and "conservatives" as is often imagined, but actually a three-way conflict. This is a very interesting finding. nishio CORRECTION: I have posted a screenshot of when the AI report was generated. The one I posted at the beginning of the thread was a screenshot of the current time, so it was a mismatch. https://pbs.twimg.com/media/Gxt22Z0aEAAqwTZ?format=jpg&name=medium#.png
nishio Arguments characteristic of the "free speech" school are as follows. https://pbs.twimg.com/media/Gxt3PxjawAEnIxX?format=jpg&name=medium#.png
nishio Anyone can continue to participate in voting and submitting comments here. The detailed report is linked from the commentary section, which is also automatically updated. I posted the above and now I'm down to two groups.
https://gyazo.com/4956922326279208d9ee0109c021bb44
nishio Interesting. This phenomenon often happens and also divides. I think it's a process of new people voting for a new tweet and spreading the word, so I'd like to see what happens after a while. 2025-08-05
https://gyazo.com/462e48df54c09034cf592c2ab10c6f22
https://gyazo.com/7eee9d1c10ae4401fdb6a80a4722ea5d
Cluster Description
o3.icon
Group A
The highest priority is placed on freedom of speech and respect for diverse opinions. Strongly rejects the use of labels and "discriminatory" labels to silence debate.
Cautious stance on AI utilization, but tolerates "transparency" as a collateral condition, rather than rejecting it completely. Concerned about legitimacy and accountability.
They are reluctant to prioritize human rights education and are wary of "imposing values".
As a whole, the group is positioned as a "free expression and prudent institutionalist."
Group B
Human rights agenda proponent. Overwhelmingly in favor of selective surnames and strengthening human rights education.
He also brings in the perspective of economy and competitiveness, and talks about linking human rights to "social and business vitality.
Negative on criticism of "excessive human rights" = strong belief that "promotion of human rights = positive."
The group can be described as progressive and pragmatic.
Group C
Strong traditional/prudent conservative values.
Strongly shared the recognition that human rights promotion can be counterproductive if taken too far.
In topics related to the institution of marriage and national sovereignty, priority is given to traditional institutions and national standards.
More alarmed by the method of decision-making (referendum) and the radical nature of the institutional restructuring than by the pros and cons of same-sex marriage itself.
The company's stance is "gradualism and emphasis on domestic standards rather than abrupt institutional change.
summary
A: "Labeling NG" and "AI must be transparent" - Emphasis on creating an environment for free discussion.
B: "Actively promote education on married couples and human rights" - liberal and pragmatic (competitive).
C: "Concerned about excessive human rightsism" and "Be cautious about changing the system" - Conservative and sovereignty oriented.
Thus, it can be read that the participants are divided into three poles: the free speech camp (A), the human rights proponents (B), and the cautious conservatives (C), each participating in the discussion under different value systems.
Voting results for the initial 30 seed comments
https://gyazo.com/53c72d0a9f117dbd55fc16c35f1ead66
https://gyazo.com/f065d4419a184b0470c20f7a747c5453
https://gyazo.com/307dfc055e26700dfb31fd845fccf590
---
2025-08-04
I decided that it would be a good idea for me to make it unofficially at this time when Polis is gaining momentum on Loose Computer Science Radio, and I have already communicated this to the supporters and Mr. Hoshi.
https://gyazo.com/2bf842a5b6fcede4db9e73003571d01c
I'm not sure if it's a bug on the Polis side or not, but the comments are not showing up and I'm wondering if it's too subtle to spread at this time, what should I do?
I've written the URL for the detailed report.
https://gyazo.com/0642f778db5b564f9895a089b7887e1a
---
This page is auto-translated from /nishio/星さんのPRでの議論のPolis using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I'm very happy to spread my thought to non-Japanese readers.