Personal Finance
Thinking Clearly About Investing in the AI Boom
By Jim Vernon, Editor, AI Intelligence International · Published 22 March 2026 · Reviewed against our editorial standards · About the author
A technology can be genuinely transformative and still be a poor investment at a given price. Railways, electricity and the internet all changed the world while destroying substantial amounts of capital along the way.
This article is about thinking structure rather than stock picks. It is not investment advice, and nothing here is a recommendation to buy or sell anything.
Key takeaways
- The technology-versus-price distinction: Being right about a technology's importance tells you nothing about returns.
- You may already be heavily exposed: Broad index funds now carry substantial weight in a small number of very large technology companies.
- Correlated career and portfolio risk: If your income depends on the technology sector and your portfolio is concentrated in it, both fail together.
- Picks and shovels, and their limits: The infrastructure argument — that suppliers profit regardless of which application wins — is sound in principle and heavily discovered in practice, which means it is priced.
The technology-versus-price distinction
Being right about a technology's importance tells you nothing about returns. Returns depend on the gap between what was already priced in and what actually happens.
When a narrative is widely accepted, the expectation is largely in the price. The question that matters is not whether AI is important but whether it is more important than the market already assumes.
Most people cannot answer that question, and recognising it is a more useful position than pretending otherwise.
You may already be heavily exposed
Broad index funds now carry substantial weight in a small number of very large technology companies. An investor who owns nothing AI-specific may still have a significant share of their portfolio tied to it.
Check the actual concentration before adding more. Deliberately overweighting something you already hold at twenty per cent is a different decision from starting from zero.
This is the most common unexamined exposure in retail portfolios at present, and it is easy to check.
Correlated career and portfolio risk
If your income depends on the technology sector and your portfolio is concentrated in it, both fail together. This is the classic error that employees at successful companies make with shareholdings.
The correction is not exotic. It is holding less of what you already depend on, and holding more cash than someone with an uncorrelated income would.
Employee equity counts as concentration. Vesting schedules make it feel different and it is not.
Picks and shovels, and their limits
The infrastructure argument — that suppliers profit regardless of which application wins — is sound in principle and heavily discovered in practice, which means it is priced.
It also assumes demand persists. Infrastructure suppliers are highly sensitive to the capital expenditure cycle, and that cycle can reverse quickly when expectations reset.
Treat it as one thesis among several rather than as a risk-free way to participate.
Position sizing over prediction
The decision within your control is how much, not which. A thesis you cannot fully evaluate deserves a position size that survives being wrong.
A workable discipline: no single conviction position above a share of the portfolio you could lose entirely without changing your plans.
Rebalance on a schedule rather than a feeling. A position that grows into a dominant weight has increased your risk regardless of how well it has performed.
Signals that a narrative is running hot
Valuations justified only by projections far beyond any current revenue, widespread use of new bespoke metrics, and companies repositioning into the theme without changing what they do.
None of these prove anything on their own and all of them have preceded both crashes and continued run-ups. They are reasons to size carefully, not to time.
The one reliable rule is that leverage during a narrative peak is what converts a drawdown into a permanent loss.
Worked example: checking real exposure
A portfolio described by its owner as diversified: 70% global index fund, 20% a technology sector fund, 10% cash.
Looking through to holdings, the largest seven technology companies accounted for roughly 19% of the index fund and roughly 55% of the sector fund — about 24% of the total portfolio in seven correlated names.
The owner also worked in software, so income and roughly a quarter of assets shared the same driver.
The response was not to sell the thesis but to stop adding to the sector fund, direct new contributions to underweighted areas, and raise cash from 10% to 18% over a year. Same view of the technology, considerably less correlated risk.
Check the correlation you already carry
If your salary depends on the technology sector, a portfolio concentrated in the same sector doubles a single bet. A downturn then hits income and savings at the same moment, which is precisely when savings are supposed to help.
Broad index funds already hold meaningful exposure to the largest technology companies, so many people are more concentrated than they realise before buying a single themed fund on top.
This is a general observation about concentration risk rather than advice about any holding. Anyone making real allocation decisions should talk to a regulated adviser who can see the whole picture.
Frequently asked questions
Is this article recommending any investment?
No. It describes a way of thinking about concentration and position sizing. Investment decisions depend on your circumstances, and a regulated adviser is the appropriate source for personal recommendations.
How much of a portfolio should be in one theme?
Small enough that being completely wrong does not change your plans. For most people that is a single-digit percentage in any specific conviction position, on top of whatever broad index exposure already exists.
Do index funds protect against a sector correction?
Partially, and less than they used to. Market-cap weighting means a concentrated market produces a concentrated index, so check the actual look-through weights rather than assuming diversification.
What is the most common mistake here?
Doubling up — working in the sector, holding employer equity and overweighting the same sector in the portfolio. Three correlated bets that feel like three separate decisions.
Are themed AI funds a shortcut to exposure?
They are also a shortcut to concentration and higher fees. Check what a themed fund actually holds against what your existing index funds already hold.