CIO Insights: Earnings season, brought to you by AI

15 minute read

The S&P 500 is wrapping up its strongest earnings season in years, led by tech and powered by AI. The question is how much further the AI story has to run, and what that signals for US and global market leadership.

In this month's CIO Insights, we worked through a sample of earnings transcripts across major global markets to better understand where AI is showing up in the numbers, what management teams are saying about the quarters ahead, and whether AI is reinforcing the case for US exceptionalism.

Key takeaways

• Q1 2026 was a standout earnings season, with tech leading the way. S&P 500 earnings grew 27.5% year-on-year – the strongest quarter since late 2021 and well above the 10-year average of 10%. Strength was broad-based – most sectors posted double-digit earnings growth. The leaders by a wide margin were the AI-exposed sectors: consumer discretionary rose more than 40%, while communication services and information technology were each up about 50%.  
• AI is still a tech story. Non-tech adoption is happening, but hasn't yet shown up in the data. The cleanest non-tech beneficiaries are the industrial names supplying the data-centre buildout, where their order books are extending into 2027 and 2028. Outside that supply chain, non-tech adopters are starting to talk about AI productivity gains, but those gains have yet to show up in their margins.
• Forward guidance for AI and tech is strong. Hyperscalers raised 2026 capex commitments again this quarter, AI chip and networking firms extended their growth outlooks, and memory suppliers are moving toward multi-year customer agreements. At the aggregate level, more S&P 500 companies have issued positive earnings guidance than negative; historically, the reverse is true.
• US equities still capture the most AI exposure, but global beneficiaries are also worth a closer look. Tech-related sectors comprise more than half of the S&P 500 – far more than other developed markets – so broad US exposure remains the simplest way to own the AI theme. That said, other global markets – South Korea and Taiwan in particular – offer concentrated exposure to AI memory and semiconductors, providing a non-US complement. China is a curious case: its indexes are similarly tech-heavy, but its tech leaders haven't yet shown the AI monetisation that has boosted their US peers.

(See our Glossary at the end for a breakdown of the terms used in this article.)

A standout quarter for equities, with tech doing the heavy lifting

The S&P 500 is wrapping up its strongest earnings season in years. As of mid-May, with more than 90% of companies reporting, Q1 2026 earnings grew about 27.5% from a year earlier, shown below in Chart 1. That was the highest growth rate since late 2021 and well above the 10-year average of around 10%.

Beneath that headline number, companies are also beating expectations by more than usual. In aggregate, reported earnings came in 16% above what analysts had forecast, more than double the two-year average of about 7%. Revenue growth painted a similarly robust picture: the S&P 500's top line grew by about 11%, the strongest reading since 2022, with positive contributions across all sectors. Profit margins hit about 14%, a record high for data going back to 1990.

Strength was broad-based, as shown in Chart 2. Most sectors posted double-digit earnings growth: financials grew more than 20%, while materials and consumer discretionary rose around 40%. The standout sectors were information technology and communication services, where earnings grew around 50%.

Three of these top-performing sectors house the names driving the AI buildout. The hyperscalers sit across them: Alphabet and Meta in communication services, Amazon in consumer discretionary, and Microsoft in information technology. The chip and equipment makers supplying them – NVIDIA, Broadcom, Micron, Applied Materials, and Lam Research – sit in information technology. Materials may also be benefiting from the AI buildout, through demand for cement, steel, copper, and other inputs used in data centres and energy infrastructure.

(For more on commodities, see CIO Insights: Unearthing opportunities in commodities.)

The above illustrates that the impact of AI is showing up most clearly in tech, where it is already driving revenue and profit growth. Whether that picture extends beyond tech, and where, is the harder question to answer from the headline data alone.

Beyond tech, earnings transcripts show AI's reach is uneven

To get a more detailed read on where AI is showing up in company results, we reviewed around 50 earnings transcripts from the Q1 2026 reporting season. The sample spans large tech and non-tech companies across the US, Europe, and Asia. It leans toward the AI value chain and the non-tech industries where AI exposure looked most direct, with a smaller set of other sectors included for contrast.

Table 1 summarises the pattern we found. Outside of the tech names, we see companies falling into three broad categories: 1) AI suppliers: those supplying demand for physical AI infrastructure, 2) AI adopters: those using AI to automate large operations, and 3) AI aware: those discussing AI as a general topic without tying it to concrete financial metrics.

1. AI suppliers: The clearest impact is in the industrial supply chain

AI-related capex is reaching industrial companies that provide power generation, electrical equipment, automation, construction equipment, and grid infrastructure. Across the industrial names in our sample, companies like Caterpillar, Siemens, and Schneider Electric point to stronger orders, longer backlog visibility, and capacity expansion plans tied to data-centre growth.

Data centres require reliable power, backup generation, grid connections, cooling, and electrical equipment, and that demand is starting to show up in industrial order books. Siemens, for example, pointed to backlog visibility into 2027, while Caterpillar said some orders now extend into 2028, and is nearly tripling its capacity for power-generation equipment. The broader picture is that AI capex is now feeding into the physical infrastructure layer that supports data-centre growth.

2. AI adopters: Spending is rising across non-tech sectors, but the margin impact is still hard to see

For companies using AI internally, the most concrete non-tech disclosures are coming from operationally-heavy industries: banking, insurance, payments, and retail. These companies are applying AI to customer service, fraud detection, claims, compliance, payments, and back-office workflows. Companies like UnitedHealth, Bank of America, JPMorgan, Visa, Mastercard, and Walmart all point to this direction, though with varying levels of detail.

The financial impact is still uneven. Some companies are reporting better productivity, faster processing, lower manual work, or higher digital engagement. Fewer are showing a margin improvement that can be cleanly attributed to AI. The early evidence points to a gradual productivity layer building across high-volume service businesses, with some gains likely to be reinvested, passed to customers, or offset by ongoing technology spend.

3. AI aware: Elsewhere, AI is part of the story, but not yet in the numbers 

For the remaining companies in our sample, AI was usually not the main earnings story. In pharma and consumer staples, management teams focused on product launches, pricing, volumes, regulation, supply chains, and consumer demand. Eli Lilly and Novo Nordisk, for example, spent most of their calls on GLP-1 drugs, while Nestlé discussed data, digital tools, automation, and simplification without giving AI-specific operating metrics.

The most concrete AI disclosures came from companies with large digital or back-office operations, where automation is easier to describe. In other sectors, companies may still be using AI, but have yet to report it as a driver of sales, margins, or guidance.

Looking ahead, company guidance shows the AI train isn't slowing down

The latest earnings season extended a pattern that has held for several quarters. As shown in Chart 3, hyperscalers raised their capex plans again, with combined 2026 spending from the largest US hyperscalers now more than $700 billion, close to double 2025 levels. Those estimates have been revised up sharply over the past 12 months: they’re more than double the estimate a year ago and 40% higher than six months ago.

Hyperscaler suppliers are also signalling multi-year demand 

This positive guidance isn’t just limited to hyperscalers. Their suppliers are pointing to longer demand visibility too, through multi-year customer agreements, order backlogs, and capacity plans that extend across a similar time horizon.

In memory, SK Hynix said customers are increasingly trying to secure medium- to long-term supply volumes as shortages persist, and Micron pointed to a similar shift after signing its first five-year strategic customer agreement.

Beyond memory, as shared above, data-centre power and electrical equipment makers like Caterpillar have order books stretching into 2027 and 2028, while AI networking and semiconductor equipment vendors are pointing to demand several years out. Put together, this suggests the AI infrastructure cycle is broadening across the supply chain, rather than being a short-lived spike.

Non-AI sectors are also guiding higher

Beyond the AI supply chain, broader S&P 500 earnings guidance is more positive than usual. As shown in Chart 4, of the 267 S&P 500 companies that have issued full-year guidance for the current fiscal year (FY 2026 or FY 2027), 58% have set EPS forecasts above where analysts had it (positive guidance) and 42% below (negative guidance)1. In most quarters, more companies issue negative guidance than positive.

In proportional terms, tech-exposed sectors lean most heavily positive, with AI-exposed names in semiconductors, networking, and the hyperscalers well represented. The bulk of positive guides by volume, however, come from non-tech sectors, where AI is not the driver.Looking at the names in our sample, the drivers vary by sector. In health care, the largest source of positive guides, the drivers are drug pipelines at the pharma names, and healthier margins at the major health insurers. In financials, firms are pointing to a strong capital markets cycle, with record or near-record quarters at the major US banks and asset managers, powered by trading, advisory, and asset flows. In consumer staples, the focus is on cost discipline and pricing power. 

The breadth of positive guidance across these sectors suggests the bullish picture in forward guidance is not resting on AI alone. Even setting AI aside, US corporations are pointing to a healthier outlook than they typically do at this point in the year.

US equities lead in AI exposure, but other global markets are worth a look

The earnings picture above raises a practical question for investors: where is the best place to own the AI theme from here, and how much of it is already captured by holding the broad US index? The answer starts with the AI exposure available through major equity indexes, and their sector composition.

Key tech-related sectors – information technology, communication services, and consumer discretionary sectors – together make up around 56% of the S&P 500, as shown in Chart 5. The same three sectors account for around 38% of the broad Japanese equity index and just 18% of Europe’s. South Korea and Taiwan sit at the high end, with these sectors accounting for 69% and 89% of their indexes, respectively.

While the US market captures the full AI stack, Korea and Taiwan offer concentrated exposure 

US equities offer exposure to the full AI stack – the hyperscalers funding the buildout, the chip companies supplying them, and the enterprise software platforms beginning to monetise AI. That exposure is much smaller in other developed markets, where the AI supply chain is concentrated in a handful of names, such as ASML in Europe and Tokyo Electron in Japan. 

In South Korea and Taiwan, information technology dominates the index, and the sector's weight has stepped up as the memory and chip rally lifted both markets. Most of that weight sits in three names tied directly to the AI value chain: in Korea, SK Hynix and Samsung supply most of the world's high-bandwidth memory, while TSMC in Taiwan is the foundry for almost every leading-edge AI chip. For investors with US exposure already in place, these markets provide ways to add to the AI theme outside the US.

We should also note that the consumer discretionary sector could overstate AI exposure in Europe and Japan, because it houses very different companies depending on the market. Among US consumer discretionary companies, Amazon and Tesla are clearly tech-related names, but in Europe the sector is dominated by luxury goods (LVMH, Richemont); in Japan, it’s autos (Toyota). Adjusting for that, the US’s lead in AI exposure is likely even wider than the headline numbers suggest.

The curious case of China

China, however, is one market where the consumer discretionary sector does map to AI. Alibaba, JD, PDD, and Meituan all sit in that sector, pushing the combined tech-related weight to around 55%, on par with the S&P 500. By that measure, the index is also structurally exposed to AI. Yet the broad China index is down 7% year-to-date in USD terms and the Hang Seng Tech Index is down 12%, even as AI-exposed indexes elsewhere have rallied.

The puzzle is why this structural AI exposure hasn't translated into performance the way it has elsewhere. Based on the most recent earnings calls from two of the largest Chinese tech names, Alibaba and Tencent, two headwinds stand out:  

1. Chip access: US export restrictions limit Chinese firms’ access to NVIDIA’s leading-edge chips, forcing them to spend heavily on proprietary chip supply. That creates a near-term capex and margin burden, even as AI revenue continues to grow. Alibaba’s cloud business, for example, accelerated to 40% year-on-year growth this quarter.
2. Consumer monetisation: Tencent's President Martin Lau said on the company’s most recent call that subscription prices and paying penetration for digital services in China sit well below levels in the West, and played down the prospects for the consumer AI subscription model in the Chinese market.

In short, exposure to China’s AI story comes with more friction than the US version. US exposure is concentrated in companies supplying and monetising the AI buildout; China's is concentrated in platforms that must work around restricted chip access, heavier capex to build proprietary chip supply, and a consumer subscription model that's harder to scale than in Western markets.

(For more on China AI, see CIO Insights: China AI – Build Now, Profits Later)

Our take: AI is keeping US equity leadership intact

US exceptionalism has been a recurring debate in markets in recent years. Viewed through the lens of AI, the latest earnings season offers a bull argument for the US market. The companies driving global AI revenues, capex commitments, and forward guidance are disproportionately American, and the sectors that house them make up more than half of the S&P 500 by weight. 

As long as AI continues to drive the earnings cycle, that weighting alone gives the broad US index a structural lead. Underpinning that view is the virtuous cycle of AI we laid out in our 2026 outlook: where AI breakthroughs translate into returns, returns attract funding, and funding fuels the next wave of breakthroughs. This earnings season tells us the first two legs of the cycle are intact. And, for now, financial conditions are still supportive of the third.

(For more on AI’s virtuous cycle, see 2026 Macro Outlook: Just the FACTs.)

For investors wanting to layer in non-US exposure, South Korea and Taiwan offer the cleanest complement, with index compositions that are even more concentrated in the AI supply chain than the S&P 500 – though be mindful that these markets have already run up sharply on the back of AI optimism. China is the open question: the AI revenue is showing up in the financial results, but whether share prices follow depends on the US relaxing chip export restrictions, and its companies finding ways to monetise AI at scale.

Within our General Investing portfolios powered by StashAway, this view is consistent with our existing positioning in US equities. Investors seeking more direct AI exposure can consider our tech-focused Thematic Portfolios, and those looking to add even more targeted sector or regional exposure can do so through Flexible Portfolios.

Authors

Stephanie Leung, Chief Investment Officer

Stephanie and her team oversee the full spectrum of investment products and portfolios offered at StashAway. She brings more than two decades of investment expertise across multiple asset classes. Prior to joining StashAway in 2020, she managed investment portfolios at institutions such as Goldman Sachs and multi-billion dollar family offices in the region.

Justin Jimenez, Head of Macro and Investment Research

Justin has more than a decade of experience in economic and investment research, and contributes to shaping the investment office's views on the global economy and asset classes. Prior to joining StashAway in 2022, he was an economist at Bloomberg.

Glossary

Earnings season

The period each quarter when listed companies report their financial results. In the US, it usually starts a few weeks after the quarter ends and runs for about six weeks.

Earnings per share (EPS)

A company's net profit divided by its number of outstanding shares. It's one of the most common ways analysts and investors measure profitability.

Profit margin

The share of a company's revenue that remains as profit after costs. Higher margins generally indicate stronger pricing power or operating efficiency.

Forward guidance

A company's own forecast for future revenue, earnings, or other financial metrics. Investors watch for guidance because it signals how management sees the quarters ahead.

Capital expenditure (capex)

The capital a company spends on long-term assets like buildings, equipment, and infrastructure.

Hyperscalers

The largest cloud computing companies, based on the scale of their data-centre operations. This group includes companies like Alphabet, Amazon, Meta, and Microsoft.

References

1. Butters, J. (2026). Earnings Insight. FactSet. Retrieved from: https://advantage.factset.com/hubfs/Website/Resources%20Section/Research%20Desk/Earnings%20Insight/EarningsInsight_052126.pdf