RFI for Artificial Intelligence Framework Services

The issuing agency is seeking artificial intelligence framework services to establish robust governance, strategy, and performance monitoring. The agency is trying to solve the challenges of AI bias, lack of system explainability, unstandardized workflows, and unmonitored system drift. The scope of work covers four main areas, starting with ethics and integrity controls. The selected vendor must develop bias mitigation protocols, create explainability guidelines, establish fairness assessment procedures, and integrate checkpoint assessments into AI system release gates. For the people, culture, and training component, the agency requires the development of role-based playbooks, golden templates for repeated workflows, and an integrated prompt library. The vendor must also deliver a workforce training tracking dashboard, monthly status updates, and quarterly leadership reports. The strategy and intake scope requires the design of standardized AI use case intake forms, evaluation criteria, a decision log of reviews, and a centralized inventory of approved use cases. To measure performance and monitor drift, the vendor must define key performance indicators, build live performance dashboards, implement drift detection mechanisms, and track return on investment. Eligible bidders must be onshore organizations based solely in the United States. The agency requires all work to be performed offsite at the vendor's own office location. The agency is looking for proposals to determine the budget for this project. The source document does not disclose specific evaluation criteria or key dates for this request for information.