Are we in an AI bubble? what it means for advisers and portfolios

Surging valuations across the so-called “Magnificent Seven” have reignited debate around whether AI-driven markets are entering bubble territory. While today’s tech giants are underpinned by strong revenues and demand, questions remain over how sustainable current growth expectations are.

In this exclusive piece for IFA Magazine, Julieth Amaya, Investment Strategist in M&G’s Life Investment Office, explores what this means for advisers assessing concentration risk, market exposure and long-term portfolio positioning.

With the “Magnificent Seven” now making up an astonishing slice of the S&P 500 representing roughly one third of the total market value, it’s no surprise people are stirring up about bubbles again. Share prices for Nvidia, Microsoft, Amazon, Google, and the rest have surged so dramatically that it’s easy to draw parallels with the late 90s tech boom.

But unlike the early 2000s when companies with little more than a “.com” name were attracting sky‑high valuations, today’s AI leaders are generating very real revenue. Their valuations are steep, yes, but they’re supported by extraordinary demand for AI hardware, cloud capacity, and services. That’s a very different foundation from speculation alone.

Still, these valuations also assume that AI adoption will keep accelerating. Any disappointment on that front could lead to a reset, especially given how heavily major indices lean on a handful of tech giants.

Where we are in the AI journey

We tend to see the AI cycle in three broad stages:

1. The Build‑Out (our current phase)

Right now, we’re deep in the investment-heavy stage. Firms are spending aggressively on data centres, chips, and the plumbing needed for AI at scale. Hardware and cloud companies are the big winners so far. This is the groundwork-laying period before adoption truly ramps up.

2. Adoption

This is when things get interesting. Personal use of AI tools has taken off, but workplace adoption is still early. Plenty of organisations are experimenting, but concerns around cost, governance, risk, and workflow disruption are slowing things down. The enthusiasm is there, it just hasn’t fully translated into business-wide integration.

3. Productivity

This is the stage everyone’s waiting for: actual productivity gains. Forecasts vary wildly, from “barely noticeable” to “economy‑changing.” Until we see it show up in company margins or broader economic data, it remains theoretical. So long as AI remains function-enhancing rather than function-replacing, its productivity effects will be harder to quantify in macroeconomic data.

The word function is used because one job could be comprised of several functions and hence productivity gains are not necessarily synonymous with job loss. 

Additional factors shaping growth

It’s important to remember that geopolitics and resource constraints are increasingly shaping the trajectory of AI growth. As the US and China compete for technological leadership, restrictions on advanced chip exports have only spurred China to accelerate domestic innovation, narrowing the gap in both hardware and software and underscoring how globally interdependent AI development has become.

At the same time, the infrastructure required to support large scale AI is straining existing systems, with soaring energy demands, chip shortages, and hyperscale data centres concentrated in the hands of a few US giants. Meeting these challenges will determine how effectively AI can scale.

What could happen from here?

The optimistic scenario:

AI adoption accelerates, capital investment stays robust, and productivity gains land at the higher end of expectations. Markets could see another leg of growth and importantly, one that spreads beyond today’s megacap tech winners.

The cautious scenario:

Adoption slows, productivity disappoints, or technical and resource constraints bite. Spending eases, earnings growth softens, and valuations compress. If AI revenue comes in meaningfully below expectations, profits could fall enough to drag on tech‑heavy indices.

What this means for investment?

Advisers are increasingly asking how AI affects investment long-term. It’s easy to get swept up in the excitement, but it’s just as important to stay diversified and avoid overconcentration in megacap US tech. Globally diversified funds like PruFund, help capture the upside of AI while avoiding megacap concentration risks through higher exposure in Asian equities. These stocks are usually cheaper to acquire and the regions’ specialisation in AI hardware benefits from AI build out demand.     

AI has reshaped how we think about the global economy, without a doubt.  But these assumptions are not prescriptive. The best approach is to stay exposed to the opportunity while maintaining the balance and discipline to navigate whatever twists and turns come next.

References

The Magnificent Seven’s Market Cap vs. the S&P 500 | The Motley Fool