CFI is the Canadian training company behind the FMVA, which has become a common credential for junior financial analysts who want something cheaper and faster than the CFA. Its AI offering started life as the AI for Finance Specialization and has now been rebadged as the AI Finance Professional, or AIFP, certification. The old URL redirects to the new one, so if you see both names around, they are the same programme. The pitch is simple and I think correct.
Most AI training teaches you how the tools work. CFI wants to teach you how to use them inside finance workflows, where the quality of the answer depends on the assumptions, on what changed since last month, and on whether anyone checked the model's numbers against the source. The course page says it directly: develop a repeatable approach to verifying AI outputs. For finance teams, that is the right emphasis.
The biggest risk with ChatGPT in an FP&A team is not that people fail to use it. It is that someone pastes an AI-generated variance commentary into a board pack without checking it. There are twelve courses, and CFI publishes a skills weighting rather than a full syllabus. AI fundamentals takes 40%, financial modelling 22%, FP&A 12%, data analysis and data visualisation 9% each, and AI governance and responsible use 8%.
In practice that means a long run-in on how LLMs work, prompting and tool choice, followed by applied work on building models and forecasts, writing commentary and producing charts with ChatGPT, Claude and the Excel copilots. The instructors are a mix of CFI's long-standing modelling people and two VPs of AI training, one of whom runs CFI's AI Labs series. The finance side is clearly in good hands. The weakest part for me is that 40% fundamentals weighting.
If you have used ChatGPT seriously at work for a year, a chunk of it will be revision. The other limit is scope. There is no Python, no API work and no automation beyond what you can do in a chat window or a spreadsheet add-in. That is a deliberate choice, since the audience is analysts who do not code, but it does mean you finish knowing how to use AI rather than how to build with it.
Where it wins is value. The AIFP is not sold on its own. It comes with a CFI membership, which costs $497 a year for Self-Study or $847 for Full-Immersion, and both prices are shown with a permanent-looking discount (around $373 and $593 when I checked). That membership also includes the FMVA, FPAP and the rest of CFI's catalogue.
If you were going to do the FMVA anyway, the AIFP is effectively free. Even on its own terms, a few hundred dollars for twelve assessed courses is cheap compared with executive education programmes charging $3,000 to $5,000 for similar business-level AI content. On the credential: CFI certifications are recognised in the sense that hiring managers in corporate finance often know the FMVA. The AIFP is new and has no track record yet.
I would treat it as proof that you put the work in, not as something employers screen for. The on-page rating of 4.9 from 109 reviews is CFI's own figure and is too small a sample to take at face value. My 4.0 is for a sensible, finance-first AI programme at a fair price, with the caveat that it stays at the user level. For an analyst who lives in Excel, I think it is the best first step in this list.