IRCC is being trained by data scientists within IRCC. What's happening here is that we need algorithm impact assessments. IRCC has begun to do that with a few of the categories, but it can't be opaque. It has to be relatable. It has to be in plain language.
I noticed that one of the recommendations was for an expert panel, and that's helpful, but ultimately we—and I mean we laypersons—have to be able to understand what is going into those rules.
I'll give you an example of a study permit. Are individuals over the age of 30 triaged in a different category because there's a deemed assumption that someone over 30 is really not pursuing education at that stage of their lives? Mr. Ghai provided some examples of other ways that themes can get triaged.
This has to be in plain sight. We have to go behind the curtain. There's no need to have a lack of transparency.