Advanced · Self-paced learning
Inquiry This course is not yet bookable.
In the course case, a system wrongly recommends rejecting an application and the recommendation is adopted in two decisions. Preserve the outputs and decision records, initiate a review of both cases and establish who informs affected people where necessary. For a restart, specify investigation of the cause, review of comparable cases and a documented test.
Set out who in the business and operations handles the incident, what evidence is needed and how a restart is decided. Use anonymised organisational information or the example case; technical remediation remains with the relevant specialists. A planned source and handover aid keeps reports, open questions and follow-up assignments together for further work.
Your incident and restart checklist and worked case record help you trace affected decisions and check the conditions for a restart.
Basic knowledge of an existing or planned AI application and its business responsibility. For your own case, access to approved information within a defined scope is sufficient; sensitive incident records are not required, and an example case is provided.
Planned self-study with 5–6 hours of work, case decisions and explanatory self-checks; no facilitated live session.
Planned learning time: 5–6 hours
