How OppIntell governs research it publishes automatically

Controls attestation

OppIntell publishes opposition research at a volume that makes per-post human review impossible, so the trust model is built into the pipeline rather than into an editorial queue. Every research claim carries a source URL, a source type (FEC filing, Secretary of State portal, court record, official campaign site, or government press release), and a citation_validation_status that is checked against the live document, not just recorded once at ingestion time. A claim whose citation stops resolving does not flow into a public post, and content that opens a research question without a record that settles it uses deliberately conservative posture language rather than a declarative claim.

Before any draft becomes a live URL, it passes a database-layer publish guard that enforces a minimum content length, a minimum number of source-backed signals per candidate-level briefing, the presence of structured data, and the absence of a broken citation or a source whose reliability score has dropped below threshold. This guard is the actual admission control for what becomes public, applied uniformly across the entire pipeline.

Corrections and privacy/removal requests are handled through an append-only ledger, not through direct edits to a live database row. Requests, applied actions, and the full request-to-resolution timeline are each recorded in their own auditable table (trust_correction_privacy_requests / trust_correction_privacy_actions / trust_correction_privacy_events), and applied-action rows are immutable once written — a request cannot be marked resolved without at least one applied action already on record.

Conflict-of-interest disclosure

OppIntell’s business model is auto-published opposition research: candidate and race research generated and published through the pipeline described above, and made available — including through paid access, exports, or licensing — to campaigns, committees, and other political clients. Where a business affiliation between OppIntell and a covered campaign, committee, PAC, or official is material to a research subject, OppIntell discloses it on the relevant page or research record rather than recusing coverage silently.

Because the same research pipeline serves every candidate and race in coverage without a human editorial decision about which candidates to cover, OppIntell does not select subjects based on which committee or campaign is a paying client. Coverage of a candidate is not conditioned on that candidate’s campaign, or an opposing campaign, being a customer.

Fair market value

OppIntell’s paid products are priced and structured to reflect the fair market value of the research and data-delivery service actually provided — structured, source-cited research assembled at the scale described above — rather than priced, discounted, or bundled in a way that could function as an in-kind political contribution or benefit to any campaign or committee.

Where an engagement could reasonably be read as providing services below fair market value, OppIntell documents the fair-market-value basis for that engagement in writing before it is delivered.