.svg)
Where does value move when output becomes cheap?
Simon Blanchette, of McGill University's Desautels Faculty of Management, calls the answer the human skills premium.
AI does not retire human skills. It relocates them somewhere harder to see, harder to measure and harder to build: the judgment to tell a sound answer from a merely plausible one, the communication to make it usable, and the coordination that turns individual speed into collective progress.
The report traces that shift through evidence from 28 sources, including the World Economic Forum, the OECD, BCG, Microsoft, and a field experiment in which 758 consultants using AI were faster on tasks inside its range and more likely to be wrong on one just outside it.
It closes on the part organizations get wrong. Human skills cannot be announced into existence, and one workshop cannot be expected to change how people behave under pressure. The distance between knowing which skills matter and making them show up when the work gets hard is what the report calls the practice gap.
For the decision makers, managers and L&D leaders whose work cannot wait for a clean forecast.
