{
    "$schema": "https://antichrist.net/zh-cn/public-content.schema.json/",
    "schema_version": "1.0.0",
    "record_type": "acn.public_content",
    "content_type": "virtual-page",
    "id": "acn:page:participatory-ai-governance",
    "slug": "participatory-ai-governance",
    "title": "Participatory AI Governance",
    "description": "A Cognitive Liberty framework for giving affected people real power over AI objectives, data, evaluation, deployment, appeals, and remedies.",
    "language": "zh-CN",
    "canonical_url": "https://antichrist.net/zh-cn/participatory-ai-governance/",
    "representations": {
        "html": "https://antichrist.net/zh-cn/participatory-ai-governance/",
        "markdown": "https://antichrist.net/zh-cn/participatory-ai-governance.md",
        "json": "https://antichrist.net/zh-cn/participatory-ai-governance.json"
    },
    "published": true,
    "modified_utc": "2026-06-21T18:12:24Z",
    "source": {
        "authority": "https://antichrist.net/zh-cn/",
        "data_file": "data/virtual-pages.json"
    },
    "content": {
        "slug": "participatory-ai-governance",
        "aliases": [
            "participatory-ai",
            "participation-in-llms",
            "community-in-the-loop",
            "democratic-ai-governance"
        ],
        "title": "Participatory AI Governance",
        "description": "A Cognitive Liberty framework for giving affected people real power over AI objectives, data, evaluation, deployment, appeals, and remedies.",
        "kicker": "Participation / AI Governance / Cognitive Liberty",
        "subtitle": "Consultation is not control. Participation matters when it can change the system.",
        "lead": "AI governance is participatory only when people can alter objectives, dataset composition, evaluation criteria, release thresholds, monitoring rules, or remedies. A listening session that cannot change those levers is advisory at best and participation-washing at worst.",
        "schema_type": "Article",
        "signals": [
            "Participation allocates power",
            "Communities shape systems",
            "No person scoring",
            "Record what changed"
        ],
        "hero_actions": [
            {
                "label": "Power of Participation",
                "target": "the-power-of-participation",
                "style": "primary"
            },
            {
                "label": "Architecture of Defiance",
                "target": "architecture-of-defiance"
            },
            {
                "label": "Public Record and Model Memory",
                "target": "public-record-model-memory"
            }
        ],
        "related": [
            "the-power-of-participation",
            "architecture-of-defiance",
            "algorithmic-exclusion-data-deserts",
            "public-record-model-memory",
            "judgment-machine",
            "research-agenda"
        ],
        "sections": [
            {
                "heading": "Participation is a distribution of epistemic authority",
                "body": [
                    "In AI, participation is not merely user feedback. It decides who can define the problem, whose language and experience count as evidence, which harms become measurable, who can challenge a deployment, and which remedies are available after failure.",
                    "The decisive test is not whether people were invited. It is whether their involvement could change an objective, dataset, benchmark, release gate, monitoring rule, or remedy."
                ],
                "callout": "If participation cannot change a consequential decision, it is consultation—not shared governance."
            },
            {
                "heading": "The participation ladder",
                "ladder": [
                    {
                        "label": "Level 01",
                        "title": "Inform",
                        "text": "The institution publishes what it plans to do. People receive information but have no decision power."
                    },
                    {
                        "label": "Level 02",
                        "title": "Consult",
                        "text": "People submit views, incidents, or preferences. The institution may ignore them without explanation."
                    },
                    {
                        "label": "Level 03",
                        "title": "Co-design",
                        "text": "Affected people help define objectives, data requirements, evaluation criteria, and deployment conditions."
                    },
                    {
                        "label": "Level 04",
                        "title": "Shared control",
                        "text": "Participants hold formal authority over release gates, audits, remedies, stewardship, or benefit allocation."
                    },
                    {
                        "label": "Level 05",
                        "title": "Contest and revise",
                        "text": "After deployment, people can inspect outcomes, appeal, obtain correction, and trigger governance revision."
                    }
                ]
            },
            {
                "heading": "Participation across the LLM lifecycle",
                "steps": [
                    {
                        "title": "Problem framing",
                        "text": "Ask affected people whether the proposed model solves the right problem, whether automation is appropriate, and which outcomes must never be optimized."
                    },
                    {
                        "title": "Stakeholder mapping",
                        "text": "Identify decision subjects, domain workers, language communities, downstream users, auditors, and people likely to bear failure costs."
                    },
                    {
                        "title": "Data sourcing and curation",
                        "text": "Document provenance, consent, language and dialect coverage, exclusions, filters, annotator context, and benefit-sharing obligations."
                    },
                    {
                        "title": "Post-training and preference aggregation",
                        "text": "Do not flatten plural publics into one hidden reward model. Record whose preferences were represented, how conflicts were handled, and what remains contested."
                    },
                    {
                        "title": "External evaluation",
                        "text": "Use independent red teams, community evaluations, subgroup testing, and open issue channels that can alter mitigations and release criteria."
                    },
                    {
                        "title": "Deployment and appeal",
                        "text": "Provide notice, monitoring, human review, incident reporting, export, correction, and remedies linked to an accountable decision owner."
                    }
                ]
            },
            {
                "heading": "Problem formulation comes before model optimization",
                "body": [
                    "Many fairness and safety failures begin before training, when institutions choose the prediction target, the unit of analysis, the definition of success, and the range of acceptable errors. Participation that begins after those choices inherits the original blind spots.",
                    "A Cognitive Liberty review therefore asks first whether the system should classify people at all, whether it can operate on events rather than identities, and whether non-automated alternatives preserve more agency."
                ]
            },
            {
                "heading": "Participatory data stewardship",
                "body": [
                    "Communities should be able to help create, validate, document, and govern the corpora that represent their language and experience. Data statements, datasheets, provenance manifests, collective stewardship vehicles, and removal pathways create the audit surface needed for meaningful control.",
                    "Contribution is not blanket consent. Public availability does not erase context, ownership, privacy, community norms, or the right to challenge exploitative reuse."
                ],
                "cards": [
                    {
                        "title": "Authority to contribute",
                        "text": "People can add missing language, context, and records through governed channels."
                    },
                    {
                        "title": "Authority to withhold private material",
                        "text": "Participation does not require surrendering private cognition or sensitive community records."
                    },
                    {
                        "title": "Authority to correct",
                        "text": "Communities can repair labels, summaries, provenance, and downstream representations."
                    },
                    {
                        "title": "Authority to benefit",
                        "text": "Participation should include credit, compensation, access, or shared governance where appropriate."
                    }
                ]
            },
            {
                "heading": "Human feedback is political aggregation",
                "body": [
                    "Preference learning can improve instruction following and safety, but it also turns a set of human judgments into a single optimization target. The central governance questions are whose preferences were collected, who translated them into labels, how disagreement was aggregated, and whether minority positions were erased.",
                    "A participatory system preserves disagreement as data rather than treating consensus as the only valid output. It supports plural profiles, disclosed policy layers, and contestable aggregation rules."
                ]
            },
            {
                "heading": "External red teaming and community evaluation",
                "body": [
                    "Independent testers, affected users, language communities, and domain experts can reveal harms missed by internal teams. Their role is meaningful only when findings produce new tests, changed mitigations, delayed release, or revised documentation.",
                    "The public record should show recommendation status: accepted, partially accepted, rejected with reasons, deferred with owner, or unresolved. This converts participation from theater into an auditable governance process."
                ]
            },
            {
                "heading": "Participation-washing",
                "body": [
                    "Participation-washing occurs when institutions borrow the language of inclusion while keeping objectives, infrastructure, timelines, release authority, and remedies centralized. Common signs include one-time workshops after product decisions are fixed, unpaid extraction of lived experience, no public response to recommendations, and no path for participants to stop or revise deployment."
                ],
                "table": {
                    "headers": [
                        "Warning sign",
                        "Why it fails",
                        "Required correction"
                    ],
                    "rows": [
                        [
                            "Consultation after design freeze",
                            "The decisive choices are already closed",
                            "Move participation upstream into problem formulation and data governance"
                        ],
                        [
                            "Feedback without response duty",
                            "The institution can ignore input silently",
                            "Publish disposition, owner, reasons, and implementation status"
                        ],
                        [
                            "Representation without authority",
                            "Participants supply legitimacy but not control",
                            "Grant formal review, release-gate, stewardship, or appeal power"
                        ],
                        [
                            "One public for one model",
                            "Plural values are forced into a false consensus",
                            "Support modular, localized, or user-selectable policy layers"
                        ],
                        [
                            "Unpaid extraction",
                            "Communities provide data and labor while firms capture value",
                            "Use compensation, credit, benefit-sharing, or cooperative terms"
                        ]
                    ]
                }
            },
            {
                "heading": "Case studies: what substantive participation can change",
                "case_studies": [
                    {
                        "status": "Open collaborative model development",
                        "title": "BigScience and BLOOM",
                        "what_happened": "A large international research workshop built a multilingual model and corpus while foregrounding ethics, law, documentation, and governance.",
                        "critics": "Open collaboration can still reproduce infrastructure and representation gaps.",
                        "supporters": "It demonstrated that model development can distribute authorship and scrutiny beyond one firm.",
                        "pressure_point": "Who controls compute, final corpus decisions, and release terms?",
                        "cognitive_concern": "Language communities need influence over representation, not merely extraction.",
                        "remedy": "Publish data governance, decision logs, contributors, exclusions, and unresolved disputes."
                    },
                    {
                        "status": "Community-led data",
                        "title": "Mozilla Common Voice and low-resource speech",
                        "what_happened": "Contributors create and validate openly available speech data across many languages and accents.",
                        "critics": "Volunteer contribution can still lack compensation or full community governance.",
                        "supporters": "It expands representational coverage and gives language communities a practical contribution channel.",
                        "pressure_point": "Who sets quality rules and downstream licenses?",
                        "cognitive_concern": "Accent and language gaps become model-access and dignity gaps.",
                        "remedy": "Pair contribution with community review, documentation, removal paths, and subgroup evaluation."
                    },
                    {
                        "status": "Regional research network",
                        "title": "Masakhane and African-language NLP",
                        "what_happened": "A distributed community builds language technology through local expertise, collaboration, and open research.",
                        "critics": "Funding and compute asymmetries remain.",
                        "supporters": "The network shifts agenda-setting toward researchers and speakers who know the languages and contexts.",
                        "pressure_point": "Can local priorities survive external funding and benchmark pressure?",
                        "cognitive_concern": "Standard-language systems can erase linguistic knowledge and rhetorical style.",
                        "remedy": "Fund local institutions, preserve data sovereignty, and measure language-specific outcomes."
                    },
                    {
                        "status": "Corporate democratic alignment experiments",
                        "title": "Collective constitutions and democratic-input pilots",
                        "what_happened": "Labs have tested public principles, deliberation, and global prototype grants for model behavior.",
                        "critics": "The company still chooses the question, aggregation method, model, and final policy.",
                        "supporters": "The experiments show that public input can alter behavior and expose preference diversity.",
                        "pressure_point": "Does the public control deployment or only advise it?",
                        "cognitive_concern": "A single corporate constitution can collapse plural values into one orthodoxy.",
                        "remedy": "Use recurring, transparent, plural, contestable governance with public recommendation ledgers."
                    }
                ]
            },
            {
                "heading": "Open models and scrutiny infrastructure",
                "body": [
                    "Open weights alone do not guarantee participatory control. Meaningful scrutiny also requires data information, training and evaluation code, model documentation, test artifacts, known limitations, and reproducible issue channels.",
                    "Transparency is the floor, not the ceiling. Communities still need time, expertise, funding, standing, and decision pathways to use what is disclosed."
                ]
            },
            {
                "heading": "Metrics that measure power, not attendance",
                "table": {
                    "headers": [
                        "Dimension",
                        "Weak metric",
                        "Stronger metric"
                    ],
                    "rows": [
                        [
                            "Representation",
                            "Number of participants",
                            "Coverage by language, region, role, and affected subgroup"
                        ],
                        [
                            "Decision influence",
                            "Comments received",
                            "Objectives, rules, tests, or release gates changed"
                        ],
                        [
                            "Accountability",
                            "Workshop completed",
                            "Recommendations publicly dispositioned with owners and dates"
                        ],
                        [
                            "Model effect",
                            "Overall benchmark score",
                            "Subgroup error, refusal, coverage, and calibration gaps"
                        ],
                        [
                            "Remedy",
                            "Support tickets closed",
                            "Appeals corrected, records restored, and harms compensated"
                        ],
                        [
                            "Durability",
                            "One event",
                            "Standing panels, recurring audits, funded intermediaries, and revision cycles"
                        ]
                    ]
                }
            },
            {
                "heading": "The FFTAC participatory-AI rule",
                "manifesto": [
                    "Judge proposed systems, not the moral worth of users.",
                    "Move participation upstream, before objectives and data are fixed.",
                    "Keep the forum internum outside the participation requirement.",
                    "Preserve plural disagreement instead of manufacturing one hidden consensus.",
                    "Publish what participant input changed and what was rejected.",
                    "Give affected people notice, appeal, correction, export, and remedy.",
                    "Do not call consultation democracy when the institution retains every consequential lever."
                ]
            },
            {
                "heading": "Research basis and claim boundary",
                "body": [
                    "The reports support a strong conclusion: participation is a governance layer that can redistribute epistemic authority when it is linked to real decisions. They do not prove that every workshop, dataset contribution, online post, or public consultation improves an AI system.",
                    "Claims about particular models, datasets, laws, or measured outcomes remain source-dependent and time-sensitive. Public pages preserve the design principles while long-term report memory retains full detail and caveats."
                ],
                "sources": [
                    "Participation governance research corpus",
                    "Participation architecture and source-preservation records"
                ]
            }
        ]
    },
    "content_sha256": "fb724d28776139e8588cc861fecb4593bcea3b1434aedb0fb3bd61f6ed21925f"
}