Why I Think AI Will Break the Labor Market Before It Fixes It
Everyone agrees AI will change work. I think the messy middle, the part between disruption and adjustment, is where the real opportunities are.
There is a version of the AI story where everything works out neatly. New technology displaces old jobs, new jobs emerge, and within a generation or two the economy is bigger and better than before. This has happened with every major technological shift in history, and there is no reason to think it will not happen again.
But that tidy narrative skips over the hard part: the transition. And I think the transition this time is going to be rougher than most people expect.
What is different this time
Previous waves of automation mostly hit manual and repetitive tasks. AI is different because it targets cognitive work: the stuff that white-collar workers assumed was safe. Writing, analysis, code, research, even basic strategic thinking. These are not fringe capabilities anymore.
I am not predicting mass unemployment tomorrow. But I do think we are going to see meaningful displacement in specific sectors faster than the economy can absorb it. Staffing firms, certain consulting models, and commoditized professional services feel most exposed.
The question is not whether it happens. It is how fast, and whether we are positioned for the awkward middle part.
For me, the investment implication is pretty straightforward: go long the companies building AI infrastructure, and be cautious about businesses that are essentially selling human cognitive labor at a premium. The tricky part is timing. Markets tend to overestimate disruption in the short term and underestimate it over five years.
I will be writing more about specific sectors and positions as I work through them. Consider this the thesis statement.