Artificial intelligence remains one of the most important long-term investment trends, but investors may need to take a more selective approach in the third quarter. Rising capital expenditure, demanding valuations, financing costs and mounting pressure to monetise AI solutions mean investors should look beyond the largest technology companies. This outlook explores how to maintain exposure to long-term growth while broadening and diversifying portfolios through high-quality defensive companies, income strategies, energy and power-grid infrastructure, and equal-weight approaches designed to reduce concentration risk.
Key Takeaways
Artificial intelligence remains a structural trend, but it is no longer a straightforward investment story. Investors may want to retain exposure to AI while distinguishing between companies building infrastructure, businesses deploying AI solutions and beneficiaries of productivity growth. Each of these groups faces different risks today, ranging from high capital expenditure and financing costs to margin pressure and the need to demonstrate a credible path to monetisation.
Lower oil prices may ease inflationary pressure, but the Federal Reserve still has limited room for manoeuvre. A framework peace agreement between the United States and Iran could reduce inflationary pressure, but persistently high services prices, rising wages, tariffs and fiscal spending suggest investors should favour pricing power, strong cash flows and solid balance sheets.
Concentration is becoming a hidden portfolio risk. Market-cap-weighted indices still provide exposure to the winners of the AI revolution, but they also make portfolio returns increasingly dependent on a small group of the largest technology companies. Equal-weight strategies, high-quality defensive stocks, income-focused solutions, and investments linked to energy and power infrastructure may help reduce this risk without giving up long-term growth potential.
The AI Story Continues, but the Easy Phase May Be Over
As Charu Chanana, Chief Investment Strategist at Saxo Bank, explains, the key question for the third quarter is no longer whether artificial intelligence will reshape the economy. What matters more is whether investors’ portfolios have become overly dependent on one version of the AI narrative: the dominance of US mega-cap technology companies, uninterrupted capital-expenditure growth, easy access to funding and market-cap-weighted indices that continue to reward the same group of leaders.
The conclusion for investors is not to abandon AI. Rather, it is to maintain exposure to the future while protecting portfolios against herd behaviour.
In practice, this means viewing AI as several distinct market segments, building a portfolio that can withstand both persistent inflation and a higher-for-longer interest-rate environment, and reducing unintended concentration in portfolios that may appear diversified at first glance but are increasingly driven by the same factors supporting mega-cap growth stocks.
The Reality of AI: Still a Structural Trend, but Tactical Risks Have Increased
The first phase of the AI investment cycle was driven by scarcity, excitement and capital spending. Investors rewarded companies linked to semiconductors, cloud computing, data centres, memory, networking, software and energy.
The next phase is likely to look different. Growth potential alone will no longer be enough. Investors will expect evidence that AI solutions are translating into revenue, margin improvement, productivity gains and attractive returns on capital expenditure. The bar is much higher than it was only a few quarters ago.
According to Bloomberg sector data as of 11 June 2026, expected earnings-per-share growth for the information technology sector within the S&P 500 is set to remain very strong: around 58% in the second quarter of 2026, 52% in the third quarter and 43% in the fourth quarter. However, such ambitious forecasts also illustrate the scale of market expectations. Companies can no longer simply grow; they must consistently exceed expectations.
Valuations and positioning raise the stakes further. Bloomberg data from 11 June 2026 show that the S&P 500 was trading at around 21.5 times expected earnings, while the information technology sector was valued at nearly 24.8 times expected earnings. This does not necessarily indicate a speculative bubble, but it does significantly reduce the margin for error. High expectations for financial results and strong investor positioning mean that even modest disappointments may trigger outsized market reactions.
There are three tactical risks worth watching.
Token Prices Could Increase Demand Sensitivity
The first wave of AI adoption was experimental. Companies tested new solutions, launched pilot projects and sought to demonstrate that they would not miss the next productivity wave.
The next wave is likely to be more disciplined. Chief financial officers will increasingly ask whether AI implementation is actually generating higher revenues, productivity improvements or lasting savings that justify the cost.
If large-scale use of AI models remains expensive, some companies may slow the pace of adoption. On the other hand, a sharp decline in token prices could increase AI usage while also undermining revenue and profitability expectations for some AI infrastructure providers.
The risk, therefore, is not that interest in artificial intelligence will disappear. A more likely scenario is that demand becomes more price-sensitive than current market valuations assume.
Financing Risk Is Becoming More Important
The development of artificial intelligence requires enormous capital investment. Building data centres means spending on semiconductors, servers, memory, energy infrastructure, cooling systems, land and grid connections.
Only a few months ago, the market expected the Federal Reserve to begin cutting interest rates relatively quickly. Today, the outlook looks different. Higher rates for longer are increasingly being discussed, while a renewed rise in inflation means that further rate hikes cannot be fully ruled out.
A higher cost of capital means more expensive financing for investment and lower present values for future earnings. This matters especially because many AI-related companies are growth stocks whose valuations are based primarily on expected cash flows in the years ahead.
The AI trend may remain structurally strong while becoming tactically vulnerable if financing costs continue to rise.
A Tactical Peak in Capital Expenditure Does Not Have to Mean an AI Market Collapse
The market does not need capital expenditure to decline in order to change its view. It may be enough for the pace of growth in investment spending to slow.
The market has rewarded aggressive AI-related capital expenditure because it confirmed management teams’ conviction in the technology’s future. However, if component prices, energy costs, financing costs and infrastructure bottlenecks continue to rise, investors may increasingly question whether expected returns justify the scale of investment.
This is a key risk for the third quarter: not an AI market collapse, but a reassessment of expectations. If investors begin to question whether high investment growth, favourable AI economics and cheap financing can all be sustained at the same time, the most crowded AI-related positions may become more vulnerable to disappointment.
Inflation Still Holds the Key: The Federal Reserve’s Safety Net Is at Risk
The situation in the Middle East remains highly fluid. One day, the market prices in the prospect of an agreement between the United States and Iran and the reopening of the Strait of Hormuz; the next, concerns return over stalled negotiations, a breakdown in the ceasefire and renewed escalation.
This makes the outlook for inflation and Federal Reserve policy less clear and increasingly two-sided.
If the region moves towards peace, lower energy prices could reduce transportation, logistics and production costs while easing pressure on household budgets. Such a scenario would support further disinflation and could improve sentiment in equity markets.
However, lower oil prices should not be mistaken for the end of inflationary risks.
The broader price environment remains mixed. Wage growth, services inflation, tariffs, changes in supply chains and fiscal spending could keep inflation above the Federal Reserve’s comfort zone even if energy becomes cheaper.
The reverse scenario remains equally plausible. If the ceasefire breaks down, nuclear talks stall, the reopening of the Strait of Hormuz is delayed or regional tensions return, oil could regain its geopolitical risk premium. That would quickly revive inflation concerns and limit the Federal Reserve’s ability to adopt a more dovish stance.
For this reason, investors should not rely too heavily on a policy safety net from the Federal Reserve. A weaker oil-price impulse may support a short-term recovery in the third quarter, but it does not guarantee a smooth return of inflation to 2% or a rapid rate-cutting cycle.
In this environment, quality remains one of the most important criteria for stock selection. Companies with pricing power, strong balance sheets, predictable earnings and cash-flow discipline appear particularly attractive. Technology continues to stand out for its earnings-growth potential, but it is also the sector facing the highest investor expectations. Industries able to sustain profitability in a less favourable macroeconomic environment may become increasingly important.
The third-quarter strategy is therefore not about reducing risk at all costs. The focus should be on building a portfolio that can withstand different scenarios: improved sentiment following a possible peace agreement, a prolonged period of high interest rates, or renewed inflationary pressure.
Concentration Risk: When Passive Investing Is No Longer Truly Diversified
The risk is not that the technology sector will stop leading markets. The greater danger is that an increasing share of investment portfolios rests on the same assumptions: AI-related capital expenditure will keep rising, the largest technology companies will maintain above-average earnings growth, financing will remain relatively accessible, valuations will be justified by results, and passive fund inflows will continue to favour the same winners.
Initial public offerings by very large companies could add another layer of risk to index investing. If major private companies such as SpaceX, Anthropic or OpenAI go public and are added to indices, passive funds may effectively become forced buyers. This could increase concentration even further and make indices more dependent on a small group of very large, growth-oriented companies.
Passive investing remains one of the most effective ways to build capital over the long term. However, it is worth remembering that today’s market-cap-weighted index may no longer offer diversification as broad as it appears. When a small group of mega-cap companies drives index performance, buying “the whole market” can in practice mean concentrated exposure to artificial intelligence, US growth stocks and momentum-based investing.
This concentration works well when markets rise. At the same time, it makes portfolios more vulnerable to AI-related disappointments, valuation corrections, weak performance from new market entrants or slower growth in investment in AI infrastructure.
This is why traditional diversification is becoming important again. The objective is not to abandon exposure to the fastest-growing parts of the market, but to reduce reliance on a single dominant source of risk.
In the current environment, three elements deserve particular attention:
- Earnings durability: companies capable of growing even in a high-interest-rate environment.
- Valuation discipline: sectors where expectations are not as demanding.
- Different risk drivers: exposures that do not depend solely on AI capital expenditure, mega-cap momentum or falling bond yields.
Portfolio construction is therefore crucial. There is no need to abandon market-cap-weighted indices, but they may be complemented by approaches that reduce the influence of the largest companies on overall investment returns.
One such approach is equal weighting. It is not a move against the technology sector. Above all, it is a way to reduce excessive concentration risk and increase exposure to companies and sectors whose performance depends on factors beyond the development of artificial intelligence.
How to Position a Portfolio for the Third Quarter: Maintain AI Exposure, but Improve Resilience
This is not the moment to eliminate risk entirely. It is, however, a good time to make conscious decisions about which risks should remain in a portfolio.
The goal is to retain exposure to long-term growth trends while reducing the risk that a single dominant investment theme determines the performance of the entire portfolio.
Stay Invested in AI, but Examine Your Exposure Carefully
There is no need to abandon AI-related investments. However, investors should stop treating AI as one homogeneous market segment.
The AI theme currently consists of several different components. Some companies build infrastructure, others use AI to improve their own businesses, and others make AI cheaper and easier to deploy. These groups may perform differently depending on whether the market is rewarding capital expenditure, productivity gains or efficiency.
In the coming quarters, these groups may behave very differently. If the market continues to reward high investment spending, infrastructure providers may perform best. However, if investors place greater emphasis on real business benefits, capital may shift towards companies that successfully use AI to increase productivity. A further decline in model and inference costs could, in turn, support companies offering solutions that improve the efficiency of the wider ecosystem.
In the third quarter, this type of selection may prove more important than broad exposure to the technology sector alone.
Build Resilience, Not Just Inflation Protection
An investment strategy for the third quarter should not rely on one macroeconomic scenario. If the peace process in the Middle East progresses, lower oil prices may improve market sentiment and increase expectations of a softer Federal Reserve stance. However, if tensions rise again, inflationary pressure could remain elevated for longer than the market currently expects.
Long-term investors should use such volatility to adjust allocations rather than chase headlines.
In practice, three steps may be worth considering.
Treat Energy as More Than a Geopolitical Hedge
Energy has an investment case if risks in the Middle East keep oil prices in a higher range, but the long-term investment thesis is broader. AI data centres, electrification, grid modernisation and rising power demand are making energy infrastructure strategically more important.
This means that oil-price declines triggered by peace-related headlines may create opportunities for investors seeking greater exposure to energy, power equipment, utilities, power-grid infrastructure and selected commodity companies.
The risk is that traditional oil companies remain heavily dependent on conditions in the oil market. For this reason, exposure to this theme may be better built more broadly by combining conventional energy with companies that benefit from long-term growth in energy demand.
Add High-Quality Defensive Companies With More Predictable Earnings
The healthcare sector is becoming increasingly attractive. It offers growth drivers independent of artificial intelligence, including demographics, medical innovation, new therapies and stable demand for healthcare services.
The sector is valued at around 17 times expected earnings, below the S&P 500, while earnings growth is improving after a weak second quarter. Consumer staples may also provide defensive characteristics. However, with valuations at around 22 times expected earnings, the sector cannot be described as clearly cheap. Entry timing remains particularly important.
Use Income and Balance-Sheet Strength
If interest rates remain elevated and the economy avoids a deeper slowdown, the financial sector may also look attractive.
As of 11 June 2026, it was valued at around 15 times expected earnings, clearly below the broader market. The key risk, however, remains a deterioration in loan-book quality if financing costs stay high for longer.
Short-term income instruments also retain a place in portfolios in the current environment. If a rapid reversal in Federal Reserve policy is unlikely, investors can continue to benefit from attractive yields while limiting the duration risk associated with longer-dated bonds.
At the same time, investors may want to focus on companies generating strong cash flows, maintaining robust balance sheets and possessing the ability to pass rising costs on to customers. Such businesses may be better positioned in an environment where inflation declines gradually and unevenly.
The objective is not to avoid risk, but to diversify it consciously. The third quarter may bring sharp swings in sentiment, ranging from optimism over improving geopolitical conditions to renewed concerns about inflation. For long-term investors, this may be a good time to broaden exposure beyond artificial intelligence and include sectors that benefit from long-term trends linked to energy transition, earnings stability and attractive income.
Use Equal-Weight Strategies to Reduce Concentration Risk
Market-cap-weighted indices remain the foundation of many investment portfolios and have delivered strong returns in recent years. At the same time, their concentration in the largest technology companies has reached levels not seen for many years.
Equal-weight strategies can provide an effective complement to this type of exposure. They increase the share of smaller companies and sectors that may benefit if market leadership begins to broaden.
Of course, such an approach may underperform traditional indices for periods when mega-cap technology companies continue to drive market gains. On the other hand, it can provide greater portfolio resilience when markets begin to reward a wider group of companies and more diversified sources of earnings growth.
An equal-weight strategy is therefore not a bet against technology. It is primarily a way to reduce the risks associated with excessive portfolio concentration and gain exposure to a broader cross-section of the economy.
Important: This material is provided for informational and educational purposes only. It does not constitute investment research, investment advice, or an offer to buy or sell financial instruments. The forecasts, opinions and market scenarios presented may not materialise, and investing involves the risk of losing part or all of the invested capital. Any investment decision should be made independently, taking into account an investor’s individual financial situation, objectives and risk tolerance, and, where appropriate, after consulting a licensed adviser.
