CapitalTime
Articles on investing and capital management, with a quantitative focus.
Exposure to Tech and A.I.
2026-10-25
Some people think that we’re in a technology and A.I. bubble. Analysts have proposed that there’s a market mania (and bubble) in semiconductors and A.I. which could end in large losses for investors.
Financial markets are complex, and there’s no way to know if this is a bubble. For all we know, this could be a perfectly healthy stock market.
However, it may still be useful to look at the A.I. exposure of portfolios. High concentration in any single sector is risky.
How much A.I. do these portfolios contain? I wanted to see how my own PRP portfolio stacks up against some popular indices. I looked at the total A.I. stock and bond exposure by adding up the holdings in each of these:
| Portfolio or Index | Type | Exposure |
|---|---|---|
| Nasdaq 100 | Equity | 41% |
| S&P 500 | Equity | 30% |
| VT (Total World) | Equity | 20% |
| PRP | Risk parity | 6% |
My own PRP portfolio has much less exposure to A.I. due to the smaller overall weight in equities. I only hold 20% in US equity, which is where almost all the exposure comes from.
VT (Total World), tracking FTSE Global All Cap Index, represents the global stock market. The overall A.I. exposure is not as bad as the S&P 500.
Methodology
While doing this analysis, I considered the following stocks and bonds to be A.I. sector: NVDA, MSFT, AMZN, GOOG, GOOGL, TSMC-Taiwan, AVGO, META, MU, AMD, Samsung, SK Hynix, SPCX.
The exposure figures shown in this post are rough approximations. It’s also worth noting that if we actually are in a tech bubble situation, the effects on asset prices could be far-reaching. If such a bubble were to burst, the damage may not be limited to just the obvious A.I. stocks.
I am not a financial professional, and this is not investment advice.
— Jem Berkes