New work appears where old tasks disappear
The jobs of 2040 are mostly not invented yet.
Every wave of automation has destroyed tasks and created others, and the created ones are easy to miss because they do not exist yet to be counted. This is not blind optimism; it is the documented pattern. The catch is timing and distribution: the new work does not always arrive for the same people, in the same place, at the same moment as the loss.
What the evidence shows
About 60% of US employment in 2018 was in job titles that did not exist in 1940. New work is continually invented, much of it pulled into being by new technology.
The WEF's 2025 outlook expects 170 million roles created against 92 million displaced by 2030, a net gain, even as the mix of skills shifts sharply.
Automating a task tends to raise demand for the complementary tasks only people can do, which is how new specialisms form around new tools.
Net positive in aggregate can still be brutal in particular. A displaced 50-year-old in a one-industry town does not get retrained into a new title automatically. The average hides the people the transition hurts.
Position yourself near where the new tasks form: at the edge of the tool, where it meets a real human need.