I had been looking for a good AI certification and this one hit the mark.
Last week, I got the PMI Certified Professional in Managing AI (PMI-CPMAI™) certification from @Project Management Institute.
AI projects behave differently enough from other projects that the usual frameworks don't fit them well. In typical projects, we agree on the deliverable up front and manage towards it. With AI, we often don't know whether the idea is even feasible until we've looked at the data, and what the model can actually do keeps shifting as we go along. That's why I wanted a certification built around a framework designed for this.
The CPMAI framework places the right emphasis on benefits realisation from the start. With every phase, based on the updated findings, the framework requires one to positively certify that the 'Business Understanding' is still accurate, and asks for an iteration to update things if it isn't. When dealing with data pipelines that change often during the project, versioned iterations, as CPMAI requires, can make a big difference.
The exam covers five domains, and my strongest results came in 'Identify Business Needs and Solutions' and 'Responsible and Trustworthy AI Efforts'. That's 'what to build' and 'how to keep it safe'. I suspect these are also the two that get the least time on most AI projects, and can result in the silent failures we often see with AI: it does what it was designed to do, but not what it was intended for.
How sure are you of the value your AI projects are actually going to deliver?
#ResponsibleAI #AIGovernance #AIProjectManagement #CPMAI