There are a lot of components working together to make what we do work, which is why you don’t really see any other product doing the same thing. We have about 30 different machine learning models working at the same time, including large language models applied to create our Pearson occupation framework of jobs, tasks, and skills.

In evaluating the demand for specific skills, for example, we use several AI methods to create our reports. We then use another model to adapt research and apply it to other countries, building out several data curves that account for different landscapes in terms of infrastructure, development, education, etc. It’s an incredibly complex process, but one that produces more accurate results.

We take an extremely scientific approach to our research. A lot of consulting companies predict future demands by hosting workshops with futurists, gathering insights from conversations with experts, or looking at previous reports. These can be helpful, but they rely largely on our assumptions about the future.

By working with AI models and undertaking large-scale data mining processes, we don’t need to make assumptions; we can make accurate predictions based on real data and scientific evidence.