In recent years, a subcommittee of the ATP's AI Technology and Assessment Committee ("ATAC") (formerly the TBA Committee), known as the Responsible AI in Assessment or RAA, has compiled an impressive library of technology and AI-related publications and guidance documents. The widespread downloads of these documents show strong interest from the assessment and learning community. As AI and technology use in assessment continues to grow, the ATAC/RAA work is critical to helping organizations learn from others, avoid pitfalls, and use AI and technology effectively to deliver assessments that are valid, fair, and reliable. 

However, while the RAA is still going strong with over 30 members and at least two new sets of guidance released each year, evidence of the guidance's effect on technology-powered assessment processes is less clear. Guidance is only as valuable as its adoption. This panel probes how and whether ATAC's work is shaping the use of technology and AI in practice — or is sitting unread after the download.

Our panelists straddle the globe. Each comes from a different geographical region of the world, a different educational background, and a different sector of the assessment community. Two are leading authors of the guidance documents, and the third has been an ATP Global member and recently serves as the senior representative for two organizations that work closely with the Ministry of Education, Japan, to advocate for the trusted and transparent use of Education Technologies and Certification Programs.
 

This panel will explore questions such as:
●    How widely — and how — are the ATAC/RAA documents and other well-known guidance documents being used in different parts of the world?
●    What should national regulators be doing regarding technology and AI in assessment?
o    Regulators: 
o    Guidelines: 
●    How can the ATP and ATAC best construct future guidance documents that are actionable and advance the principles of transparency, trustworthiness, and fairness?
●    Should the ATP also develop a repository of use cases detailing successful and less successful AI deployments in assessment?
●    What more can the community and/or ATP do to make guidance timely, deployable, and useful?