Artificial intelligence is becoming embedded across the full spectrum of knowledge work. It now shapes how professionals research, draft, analyse, decide, coordinate, and govern. Yet while adoption has accelerated rapidly, the language used to describe capability remains fragmented. Employers ask whether people are AI-ready, individuals try to signal that they are, and sectors are beginning to define their own expectations. What is still missing is a common framework that explains what AI proficiency actually means.The CFTE AI Proficiency Framework addresses that gap. It provides a shared reference model for defining, assessing, and developing AI proficiency across the professional workforce. It bridges the market-facing language of readiness with a more rigorous concept of proficiency. The framework is built around three public proficiency levels, ten capability domains, and three assessment dimensions: knowledge, skills, and behaviours. Its public structure is supported by a more granular six-band developmental model used in diagnostics and interpretation. This allows the framework to remain simple enough to communicate and rich enough to measure.The framework distinguishes durable proficiency from time-sensitive tool fluency and from the role-dependent challenge of applied capability. It is designed to be platform-agnostic, portable across institutions and sectors, and open to adaptation with attribution to CFTE. As a parent framework, it provides the common core on which methodology, diagnostics, repository examples, and official sector profiles can be built over time. It should be understood as a foundational model rather than a final doctrine: stable enough to cite and build on, yet open enough to evolve as AI tools, professional expectations, and sector needs continue to change.
Trieu et al. (Sun,) studied this question.