AI and public content

AI & Public Content

Last updated: September 17, 2026

SourceBoard publishes public information in forms that can be read by people and, in some cases, understood by search engines, crawlers, research tools, archival systems, and artificial-intelligence systems.

This document explains the distinction between public content and private account data, and how automated systems should interpret SourceBoard information.

1. Public content is publicly accessible

Content intentionally published as public may be accessed by people who are not signed in.

Depending on the page and feature, public content may include:

• posts;
• titles;
• public comments;
• public media;
• public profiles;
• source links;
• Accepted Source status;
• Verified Source status;
• public attribution;
• public timestamps;
• public reputation or achievements; and
• other information explicitly shown publicly.

2. Automated access does not change privacy status

The fact that an automated system can read a public page does not make private SourceBoard data public.

Private account data, restricted content, hidden moderation information, authentication material, and other non-public information are not intended to be exposed merely for AI, crawler, indexing, or research use.

3. Source labels must preserve meaning

Automated systems should not flatten SourceBoard's source states into a single claim of truth.

In particular:

• an ordinary comment is not automatically a source;
• an Accepted Source reflects a post-author selection;
• a Verified Source reflects a separate Source Integrity determination;
• a disputed or revoked source should not be represented as currently verified; and
• source labels do not transfer copyright or guarantee absolute factual correctness.

4. Context matters

AI systems, search engines, and summarizers should preserve important context, including:

• whether a statement came from a user;
• whether a source is accepted, verified, disputed, revoked, or unresolved;
• whether a claim is speculation;
• whether the underlying source is external;
• whether a post has changed over time; and
• whether attribution concerns authorship, publication, or another provenance question.

5. Public does not mean unrestricted ownership

Making content public does not transfer ownership to crawlers, AI systems, search engines, or other third parties.

Third parties remain responsible for complying with applicable copyright, database, privacy, contract, and other legal requirements.

6. Indexing and summarization

SourceBoard may permit or facilitate ordinary indexing, previews, snippets, summaries, citations, and machine-readable understanding of public content.

SourceBoard may also publish technical instructions that limit, permit, or structure automated access.

7. Automated training and model development

Whether a third party may use public SourceBoard content for training, fine-tuning, retrieval, benchmarking, or model development depends on applicable law, the third party's legal basis, applicable SourceBoard restrictions, and rights held by the relevant content owners.

Public availability alone should not be interpreted as a universal license for every downstream use.

8. Attribution by automated systems

Where an automated system cites SourceBoard, it should, where practical:

• identify SourceBoard as the platform;
• link to the relevant public page;
• distinguish the SourceBoard user from the external source being identified;
• avoid implying SourceBoard ownership of third-party content;
• preserve Accepted versus Verified distinctions; and
• avoid attributing a user claim to SourceBoard as an institutional statement.

9. Hallucinations and generated claims

SourceBoard does not control outputs generated by independent AI systems.

If an AI system incorrectly states that SourceBoard verified something that SourceBoard did not verify, that statement is not an official SourceBoard determination.

Users should consult the current SourceBoard page and status history where accuracy matters.

10. Robots, rate limits, and technical restrictions

Automated systems should respect:

• access controls;
• rate limits;
• published crawler instructions;
• authentication boundaries;
• visibility restrictions; and
• anti-abuse measures.

Attempting to obtain private data through crawling, evasion, or circumvention violates SourceBoard's Acceptable Use Policy.

11. Research use

Researchers may analyze public SourceBoard content subject to:

• applicable law;
• ethical research practices;
• privacy considerations;
• SourceBoard access restrictions;
• security requirements; and
• intellectual-property rights.

Research publication should avoid exposing non-public personal information or creating misleading claims about SourceBoard verification.

12. Journalistic use

Journalists may reference public SourceBoard content under ordinary legal and professional standards.

SourceBoard recommends distinguishing among:

• what a user claimed;
• what a post author accepted;
• what Source Integrity verified;
• what the external source itself states; and
• what the journalist independently confirmed.

13. Corrections and updates

Because source information can change, automated systems should prefer current pages over stale snapshots when representing current SourceBoard status.

An archived page may show a source state that has since been disputed, revoked, replaced, or verified.

14. Contact

Questions about automated use of public SourceBoard content should be directed through SourceBoard's official support, legal, research, or contact channels.