Transparency notice

How this CopeCheck site uses AI

Scores, verdicts, commentary and some summaries may be generated by artificial intelligence. They are automated analytical outputs, not independent human fact-checks. Linked source material and public datasets remain the evidence to consult.

Models

The CopeCheck network uses OpenAI Codex/GPT-family models—currently gpt-5.6-luna with maximum reasoning for the main scoring path—and MiniMax models, including MiniMax-M3 or M2.7, for some specialist pipelines and fallbacks. Models can change as the system is maintained; the methodological tests remain the controlling standard.

Automated scoring methodology

The pipeline extracts an attributable claim, compares it with linked public evidence and scores it against CopeCheck's published structural tests: productive participation, unit-cost collision, interface collapse, propagation blindness and coordination feasibility. Automated validation checks required fields, attribution, confidence and review status before publication.

Limits

Model outputs may misread a source, omit context, make unsupported inferences or produce an incorrect score. A score is an analytical judgement, not a statement that every sentence in an underlying source is false.

Corrections

To report an error, send the page URL, disputed text and supporting evidence to [email protected].