UAO Fiduciary · Systemic Risk Radar
Diversification is supposed to protect the universal owner. A handful of AI giants have quietly undone it.
The universal owner's defining advantage is breadth: hold everything, and idiosyncratic shocks wash out. That advantage depends on the market actually being broad. Over the past few years it has narrowed sharply, as a small cluster of companies tied to artificial intelligence has come to dominate the value and the momentum of global equity indices.
For a passive, diversified owner this is a structural problem hiding inside an index. You can hold thousands of names and still find that a handful of correlated mega-caps drive the bulk of your equity risk. The label says diversified; the exposure says concentrated.
Concentration is a systemic risk, not a sector view
It would be a category error to treat this as a stock-picking question. The universal owner is not deciding whether to own the AI leaders — it owns them by construction. The risk is that the breadth which justifies the whole strategy has thinned, so that a single thesis repricing could move the entire portfolio in a way diversification was supposed to prevent.
The label says diversified. The exposure says concentrated. That gap is the risk.
There is a second-order exposure beneath the first. The same concentration runs through supply chains, energy demand and capital spending, so the correlation is deeper than the index weights suggest. The diversified owner is exposed to the AI build-out not only through the obvious names but through the power, the chips and the data centres they all depend on.
What an owner can actually do
Selling is not the tool — the universal owner cannot meaningfully underweight the market it is defined by. The tools are stewardship and measurement: pressing for governance and capital discipline at the dominant firms, and modelling concentration explicitly so the board understands that today's beta is unusually dependent on a single narrative.
The Diversification Limit, a recurring data feature in this section, exists to make that visible — to show, period by period, where the protection of breadth stops working. AI is the clearest current example of an old lesson: diversification is only as real as the market is wide.
The counter-case · the strongest opposing view
Concentration alarm has a respectable rebuttal. Index concentration is not new — the 1960s Nifty Fifty and 2000 telecoms rhymed with today — and the dominant AI names lead because their cash flows are real, not speculative. A passive universal owner is, by design, agnostic: capturing the winners is the point, and “de-concentrating” is active management in disguise, with its own risk of underperformance. The honest uncertainty is whether today's concentration reflects durable economics or a repricing waiting to happen — and confident answers in either direction are themselves the risk.
UAO Fiduciary sets out the argument and the strongest counter-argument so allocators can weigh the evidence themselves. We report the debate; we do not pick a side.
Systemic Risk Radar — The defining feature of the universal owner: the risks you cannot sell out of, and the playbook for protecting market-wide beta. · Weekly. Part of UAO Fiduciary.
Researched and edited by the UAO editorial desk.