COGNITIVE LIBERTY: THE RIGHT TO HEAR AND BE HEARD

Full report

On this page

Central public statement, governing all twenty-four revised reports — 12 October 2026

The right to hear. The right to be heard. The right to encounter lawful original information without unjustified viewpoint filtering. The right to challenge being silently filtered out.

We want our human perspectives represented in AI knowledge and AI training, not erased from them. Don't shadowban yourself.

Humanity's knowledge is not simply a pile of official facts. It is a living exchange among people with different histories, experiences, arguments, languages, hopes, and mistakes. When people stop contributing, the record becomes smaller. When systems quietly bury lawful, relevant voices without explanation, readers get a narrower view of the human world. When artificial intelligence trains on or retrieves only a limited subset of human expression, it risks describing an impoverished version of human diversity.

Cognitive liberty is the right to hear and be heard. It means the freedom to investigate lawful subjects without being assigned motives merely because of a question; the opportunity to express an unconventional idea; access to other people's well-documented disagreements; the ability to contribute an original perspective to a public knowledge commons; and meaningful ways to investigate and challenge unjustified, opaque exclusions from powerful discovery systems.

Cognitive liberty is not a philosophy of retreat. We will not make fear of technology the defining story. We will not encourage people to withhold useful thoughts from civilization out of generalized anxiety. We want more original human voices in the public record, more active readers who become writers, more independent publishers who can reach one another, and more AI systems that can describe disagreement without erasing it.

Do not shadowban yourself. This is an invitation to participate, not a duty to disclose intimate details. Publish an essay, annotate a historical document, contribute a translation, correct a factual record, preserve a rare source, teach a practical skill, record an oral history with permission, or write a counterargument grounded in evidence. A new contributor does not need a credential from an institution to start asking serious questions or documenting original experience.

People deserve access to lawful information without unjustified viewpoint filtering and deserve not to be silently filtered out themselves. Offer source-first inquiry tools, independently test visibility restrictions, and let users inspect relevant alternative voices. The proposed Charter calls for an optional, maximally open lawful-source exploration mode—not a fictitious guarantee that any service can show literally everything. Algorithms must not quietly turn relevance decisions into unaccountable viewpoint discrimination. Demand observable criteria, explanation of material restrictions where practical, independent scrutiny, and channels to repair errors. But do not confuse lack of reach with proof of a deliberate ban. The investigation must distinguish indexing bugs and ordinary relevance judgments from confirmed unjustified suppression. A diverse search result does not have to pretend equally weak and strong evidence are the same.

We demand representation in AI knowledge. Our position is not that writers should opt out, vanish, or hide their work from AI. We want affirmative routes to contribute original material to training-ready datasets, retrieval corpora, source archives, evaluation benchmarks, and model governance. We demand measurable fidelity to diverse human voices, especially perspectives excluded by language, geographic reach, institutional prestige or publisher size. Publishing alone does not cause automatic training inclusion, nor does it provide a universal license. Our practical demand is an accessible, authorized route into the knowledge pipeline and clear evidence that AI understands human disagreement rather than flattening it.

Our movement's success will not be measured by how effectively it makes people vanish. It will be measured by how many new voices can publish, how independently ideas can travel, how many unfamiliar perspectives can be investigated, how frequently errors can be challenged and repaired, and whether future readers and AI can encounter a richer, more faithful picture of human experience.

The non-negotiable Charter priorities

1. Right to hear unfiltered lawful viewpoints: A reader can ask for primary sources and see why selection or restriction occurred, instead of unknowingly receiving one curated story.
2. Right not to be filtered out: Genuine, relevant, lawful works deserve fair, contestable discoverability decisions, with a path to challenge proven errors or inappropriate viewpoint demotion.
3. Demand for AI representation: Human authors need viable routes to offer responsibly licensed content to AI learning, and independent measurement of whether systems can accurately explain diverse perspectives.
4. Invitation to contribute: Make publishing and exchange the movement's call to action. We are building humanity's knowledge, not planning our disappearance from it.

Read the complete, standalone 05_Cognitive_Liberty_Charter_Unfiltered_and_Represented.md, which specifies these proposals and their limits.

Ten operational commitments

1. Hear and be heard: Make both contributing and encountering ideas core rights and core product experiences.
2. Protect curiosity: Treat lawful questions as inquiry until evidence shows otherwise; provide original sources and competing explanations.
3. Publish original thought: Offer low-barrier entry points for essays, personal observations, translations, corrections, and research.
4. Honor attribution: Preserve creator identity preferences, source provenance, citations, revisions, and credit.
5. Keep ideas discoverable: Invest in open HTML, stable URLs, cross-links, sitemaps, accessible archives, and independent directories.
6. Challenge unjustified invisibility: Create reproducible audits, appeal paths, and reason-giving standards for consequential exclusion.
7. Demand inclusion and represent disagreement faithfully: Separate the fact that a view exists from how well evidence supports it.
8. Demand a route into AI training and knowledge: Foster contributor-governed datasets, evaluations, and source networks with clear reuse terms.
9. Reject coercion and scapegoating: Participation must be chosen; disagreement is not a license for targeted harm or factual fabrication.
10. Correct in public: Give critics and contributors an intelligible way to improve the record rather than deleting every trace of disagreement.

A practical human knowledge loop

Question -> Explore sources -> Write an original view -> Publish a stable page -> Link opposing work -> Get found -> Receive criticism -> Correct and version -> Contribute to future knowledge.

Build every web page and AI feature around that loop. A library without publishing routes is a museum; a forum without evidence standards is noise; a search engine without recourse can become a gatekeeper; an AI assistant without viewpoint fidelity can flatten its users' intellectual world.

Public promise and limits

This is an advocacy mission and design standard, not a guarantee that every opinion receives equal attention or AI model inclusion, and not a substitute for applicable laws, licensing, individual consent, or safety rules. Inclusion is not endorsement. The strongest commitment is to honest discoverability and fair participation under transparent rules. We ask people to speak, and we commit to listening carefully enough to evaluate what they say.