4 min read
The Return of the Generalist
For thirty years the advice was to specialize. AI is quietly reversing that. The person who can do a little of everything competently is becoming more valuable than the person who does one thing perfectly.
The advice my generation grew up on was to specialize. Pick a narrow thing, go deep, become the person nobody can replace because nobody else understands the corner you own. It was good advice for the world it was given in, where depth was scarce and the cost of switching contexts was high. I think AI is in the process of quietly turning that advice on its head, and the people who notice early are going to do well.
What a model does best is give you competent depth in a domain you do not personally know. It writes a passable first draft of the legal language, the marketing copy, the SQL query, the design brief. Not expert work, but the work of a junior who knows the basics. If competent depth in any single area is suddenly cheap and available on demand, the scarce and valuable thing shifts to the ability to move across areas and tie them together. That is the generalist, and the generalist is coming back.
What the model is good at is exactly what specialists sold
A specialist's value used to come from holding deep knowledge that was expensive for anyone else to acquire. The corporate lawyer knew the contract patterns. The analyst knew the query syntax. The designer knew the conventions. You paid them because the knowledge was locked up in years of training and you could not get it any other way.
A lot of that knowledge is now a question away. The model will not replace the top specialist, the one whose judgment is genuinely rare. But it absolutely competes with the median specialist, the one whose value was knowing the standard moves in a standard domain. When the standard moves are free, being the person who only knows the standard moves in one domain is a weaker position than it has ever been.
The generalist becomes the one who steers
Here is what gets more valuable in that world. The person who understands enough about five domains to know what to ask for, to recognize when an answer is wrong, and to assemble pieces from different areas into something that works. The model can produce the parts. Someone has to know which parts to ask for and how they fit, and that someone needs breadth more than depth.
I watch this in my own work constantly. The most useful person on a project is rarely the deepest specialist. It is the one who understands the business problem and the data and the interface and the constraints well enough to direct the work, even though they could not personally do any single piece as well as a dedicated expert. The model handles the depth. The generalist provides the direction, and direction is the thing that does not come out of a model on its own.
This is good news for a lot of people
The specialization era was hard on people who were curious about many things and were told that curiosity was a liability, that they needed to pick a lane and stay in it. The skills that era undervalued, breadth, synthesis, knowing a little about a lot, judgment about which questions matter, are exactly the skills that pair well with a tool that supplies depth on demand. A lot of people who felt they did not fit the specialize or die advice are about to find their shape is suddenly the useful one.
It is also good news for how careers can be built. The old path meant committing early to a narrow track and hoping it stayed valuable. The emerging path rewards the person who keeps learning the surface of new domains, because each new domain they can navigate is another place they can direct a model that already knows the depth. Breadth compounds in a way that a single deep specialty does not when the specialty can be rented by the hour.
What to actually do with this
If you are early in a career, resist the pressure to over specialize before you have to. Get genuinely competent at directing work across a few domains rather than perfect at one. Learn enough of the depth to tell good output from bad, because the generalist who cannot recognize a wrong answer is worse than useless, but do not mistake depth in one area for the only path to being valuable.
If you are hiring, look harder at the people who do not fit a single box, the ones who have a credible story across several areas. The instinct to hire the deepest specialist made sense when depth was scarce. Now that depth is cheap and synthesis is rare, the person who can hold the whole problem in their head and direct the pieces is the one who makes a team faster.
The future of work conversation is mostly about loss, about which jobs disappear. This is one of the corners where something is being created instead. The generalist spent thirty years being told to narrow down. The tools just made breadth valuable again, and the people who kept their curiosity alive through the specialization years are about to be glad they did.
Let's Connect
8939 South Sepulveda Boulevard Suite 102
Los Angeles CA 90045
United States