The Blackwell-Hart Technical Bulletins
The Blackwell-Hart Technical Bulletins serve as the research and commentary archive for the Blackwell-Hart Methodology™ (BHM™). The archive documents applied observations, case studies, and structured analysis related to AI-assisted discovery systems, authority infrastructure, and entity interpretation.
Blackwell-Hart Methodology™ Technical Bulletin 26-31: Relevance Is Not Recommendation — When an Entity Fits the Question but Is Not Selected
An entity can be relevant to a question without being selected for recommendation. Blackwell-Hart Methodology™ (BHM™) distinguishes retrieval, recognition, interpretation, relevance, and recommendation as separate observable stages, providing a more precise basis for evaluating how AI-assisted discovery systems select entities under defined query conditions.
Blackwell-Hart Methodology™ Technical Bulletin 26-29: Baseline Integrity — Why Change Cannot Be Measured Without a Starting Point
As AI systems move from generating answers to taking actions, accountability depends on more than the model itself. Examining the Australian Medicare incident, this article explores access permissions, safeguards, monitoring, incident reporting and evidence-based accountability through the Blackwell-Hart Methodology™ perspective.
When AI Knows Who You Are but Doesn't Recommend You
AI systems may recognize an entity when named but omit it from answers to relevant, non-branded questions. This experiment examines entity resolution, source relationships, category association, and recommendation across ChatGPT and Gemini.