What it means
Technical feasibility is the assessment of whether a proposed AI solution can realistically be built and operated with the available technology, data, systems, skills, budget and constraints. It considers the full system, not just whether an AI model can perform the task.
Real-world example
A tax office wants AI to extract data from incoming paper letters. The model performs well in tests, but the feasibility assessment finds that letters are currently scanned at a resolution too low for reliable OCR, the case system has no API for writing data back, and the hosting must be approved for tax data. Each finding becomes a work package or a reason to rethink.
Why it matters for AI adoption
Feasibility assessments catch problems early, when changing course is cheap. They should include IT operations, security and data protection, not only data scientists, because integration and operating constraints are often decisive.
Questions to ask
- Beyond the AI model, what would we need to build, integrate and operate?
- Who from IT operations, security and data protection has reviewed this?