To tell whether a portfolio is overexposed to AI stocks, add up the exposure from shares you own directly and the same companies held inside your ETFs and mutual funds. Then compare the combined weights—and the risks they share—with your own target allocation, time horizon, and tolerance for losses. There is no universal AI-stock percentage that makes a portfolio overexposed.
How to measure AI exposure across your portfolio
Start with a clear scope. Decide whether you are reviewing one account or all investment accounts, and use market values from the same date. If you leave out retirement accounts, employer shares, cash, bonds, or other assets, the result is only a partial picture. Keep the denominator consistent: each exposure should be expressed as a share of the same portfolio total.
As an Amazon Associate I earn from qualifying purchases.
- List direct holdings. Record the market value of every stock you own directly, including employer stock and shares held in different accounts.
- Look through funds. For each ETF or mutual fund, check its latest available holdings disclosure, prospectus, or fund website. Record the fund’s portfolio weight and the weight of each relevant company inside it. FINRA recommends checking these underlying holdings because the same stock can appear directly and inside both a technology fund and an index fund: Concentrate on Concentration Risk.
- Choose and state an AI-stock definition. “AI stock” is not a standardized category in the cited investor and index materials. Name the companies or data-provider category you include, and date the list. A technology company, semiconductor maker, cloud provider, or member of the Magnificent Seven does not automatically have the same degree or kind of AI exposure as another.
- Aggregate overlapping exposure. For a fund that makes up fraction F of your portfolio, and a company that makes up fraction W of that fund, the company’s indirect portfolio contribution is F × W. Add that contribution across every fund, then add any directly held shares of the same company. Repeat for each company on your chosen list.
- Compare both issuer and theme totals. Add the look-through weights for the companies in your AI list to estimate the theme’s share of the portfolio, but keep each company’s combined weight visible. A theme total can conceal an outsized position in one issuer.
Record the fund name, its portfolio weight, each relevant underlying company weight, the resulting look-through contribution, the combined company exposure, and the source and date of the fund holdings. Fund disclosures change, so an estimate is only as current as the holdings data behind it.
Why many holdings can still mean concentrated exposure
A ticker count is not a diversification test. Several funds may own the same large companies, and a broad market-cap-weighted index fund can carry meaningful indirect exposure to them. FINRA puts it plainly: “Simply holding only funds doesn’t shield you from concentration risk.”
#1 Best Overall
Look beyond company names to the factors that could affect them together. Two positions in different industries may still depend on the same data-center spending cycle, financing conditions, or expectations for AI revenue. Conversely, a company’s inclusion on a broad AI-related list does not establish that its business or returns are driven by AI in the same way as every other company on that list.
- Issuer weight: How much of the total portfolio depends on each company after adding direct and fund exposure?
- Sector, industry, and theme: Do several positions rely on the same business activity or spending trend?
- Geography and company size: Are positions clustered in a particular market or in a small group of very large firms?
- Overlap and return drivers: Which funds share underlying names, and what common conditions could move those names together?
- Liquidity and allocation fit: Could you sell or rebalance when needed, and does the combined exposure fit the role this part of the portfolio is meant to play?
S&P Global Market Intelligence’s August 25, 2026 analysis discusses stress testing against the Magnificent Seven as a correlated basket. It also cautions that measured historical correlations depend on the lookback period, return frequency, and weighting choices; past co-movement cannot promise future behavior. Correlation is one lens, not a complete retail risk score or a forecast of safety: Using Implied Correlations to Stress Test and Diversify AI Concentration Risk.
What index statistics can—and cannot—tell you
Published figures show why it is useful to examine definitions and dates, but they are not limits for an individual portfolio. Index composition, a fund’s holdings, index performance contribution, and your own look-through exposure are different measurements.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11| Published observation | What it describes | How to interpret it |
|---|---|---|
| Average top-10 weight of 37% across seven selected AI-focused indices, compared with 78% for the S&P 500 Information Technology Index | ESMA’s analysis of selected index compositions in its report dated February 25, 2025 | This is not a figure for every AI fund or a recommended household allocation. |
| 115 firms (58%) appeared in only one of the seven indices; 16 constituents appeared in at least five | ESMA’s comparison of those indices in the same 2025 report | The selected providers’ AI universes differed; “AI stock” depends partly on the chosen definition. |
| Magnificent Seven accounted for 50% of the S&P 500’s year-to-date gain as of October 2024 | An October 2024 performance observation described in ESMA’s 2025 report | It measures contribution to index gains, not portfolio weight or an investor’s AI exposure. |
| At mid-2024, the Magnificent Seven had nearly one third of S&P 500 market capitalization and nearly 23% of MSCI World; their combined S&P 500 weight had more than doubled over the preceding ten years | ESMA’s dated index-weight context in its February 2025 report | These are mid-2024 figures, not current index weights or a personal portfolio threshold. |
| A basket of NVDA, MSFT, AMZN, META, AVGO, GOOGL, ORCL, and AMD drove more than half of S&P 500 returns and almost one third of global equity returns in 2025, as of October 28 | Invesco’s 2026 outlook, using its own AI-name basket and data through October 28, 2025 | This is a source-specific performance contribution, not a measure of all AI exposure. |
| Data-center investment contributed 1.1 percentage points to U.S. GDP growth in the first half of 2025 | Invesco’s 2026 outlook; the report notes that the figure ignores a partially offsetting import component | This is economic context, not a measure of stock or portfolio exposure. |
Sources: ESMA, Artificial intelligence in EU investment funds; Invesco, 2026 Investment Outlook: Resilience and Rebalancing.
Compare the result with your allocation plan
A concentration flag is a reason to review your plan, not an automatic sell signal. FINRA says asset allocation depends on factors including risk tolerance and investment horizon, and recommends periodic review. Its guidance does not set a universal rebalancing timetable or an AI-specific cutoff. Compare the look-through result with the allocation you intended and consider whether it still suits your circumstances: Asset Allocation and Diversification.
Concentration can be deliberate, but it can also build without a new purchase: a position may grow after strong performance, employer stock may accumulate, or several investments may become more dependent on the same risk drivers. Include liquidity in the review, especially if a large holding may be difficult to sell when needed.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Ways to respond—and costs to consider
If the exposure no longer fits your plan, possible rebalancing approaches include directing new contributions toward underweighted parts of the portfolio, redirecting available cash, or selling some of an overweight holding. The appropriate choice depends on the account, investment, and your circumstances.
- Sales may involve transaction charges or other fees.
- Selling in a taxable account can have tax consequences; rules vary by jurisdiction and account type.
- Selling after a decline can lock in losses, so consider the reason for the change rather than reacting only to a price move.
- Complex holdings or exposure that is difficult to aggregate may merit review with a qualified financial professional.
No personal holdings, fund weights, or target allocation are available here, so this method cannot determine whether any particular reader’s portfolio is overexposed. It is general educational information, not individualized financial, tax, or investment advice.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
