Hong Kong links AI training to school plans and evaluation criteria

School-system leaders and professional-learning teams need training that produces an executable school plan. Ask which decisions participants must make afterward, who can allocate staff time, and what evidence determines continuation. A bounded engagement could help one school team develop a use-case plan with baseline, owner and review date. The value hypothesis is improved readiness to evaluate implementation. Do not present Hong Kong's announced program as proven effective, as a U.S. obligation, or as an available local procurement opportunity.

Build a training sandbox around approved local tasks and synthetic records. Prerequisites include curriculum objectives, accessible devices, permitted model access and a support route. Separate a demonstration agent from any system able to change records. Validate access restrictions, audit capture and fallback before exercises. A proposed proof of value is a complete data-flow and evaluation plan for each selected use, reviewed by district specialists. This circular gives no basis for hardware sizing or a preferred hosting architecture.

The school principal should sponsor a team with curriculum and technology representation. Secure release time and train participants to write observable success criteria and record failures. Dependencies include local data approvals and a baseline workflow.

Proposed acceptance

the team produces an approved pilot plan, names its operating owner, completes a rehearsal and schedules an outcome review before expansion. Risks include attendance becoming the only metric, staff turnover breaking ownership and borrowing international assumptions without local adaptation.