AI Market Trends: Q2 2026 Review and Forecast

Q1 2026 brought a seismic shift in the AI market, moving from theoretical potential to tangible capacity. As capex raises concerns, the focus is now on AI's return, autonomous systems, and the race for power and memory. How will this reshape the tech landscape?

Explore AI market trends with insights on semiconductor growth, memory price hikes and upcoming IPOs. Artificial intelligence (AI) equities continue to be defined by volatility, with the Nasdaq-100 (INDEXNASDAQ:NDX) giving back some of its earlier momentum in the final days of the second quarter. Against that backdrop, Jefferies predicts that memory prices will rise 40 to 50 percent in Q3, plus another 30 to 40 percent in Q4; a potential 40 to 45 percent increase is on deck for 2027.

Meanwhile, as the physical economy buildout continues, attention is shifting toward the evolving funding mechanisms supporting massive AI projects. To find out more about what's coming for the AI market in Q3 and beyond, the Investing News Network spoke with Nicholas Mersch of Purpose Investments, honing in on how the anticipated entry of major application-layer companies into the public markets promises to challenge the current dominance of infrastructure-focused returns. Semiconductors, mega-CAPEX and the next wave of AI-driven returns Over the next six to 12 months, equity markets are likely to remain dominated by a narrow but powerful set of themes.

Despite bouts of volatility — the PHLX Semiconductor Sector (INDEXNASDAQ:SOX) lost more than 10 percent on June 5 on concerns about a potential stock market correction — Mersch said that in his view semiconductors will continue to be the core of the AI trade, and an arguably underappreciated area. “If you go back just to the beginning of the year, up until March 31 SOX was essentially flat. Now, it’s up (again).

The entire market is saying, all this CAPEX is being spent on the AI ecosystem overall, so where is that going? Who are the beneficiaries of that? It’s all going to the semiconductor stocks," he said.

Mersch argued that the AI capex surge has effectively “pulled earnings forward” for key chipmakers, making them look less like distant, long-duration growth bets and more like near-term cash generators. He named Micron Technology (NASDAQ:MU) as a prime example of this idea, noting that while the company's share price is up roughly tenfold, its earnings have increased closer to twelvefold, a gap that, in his view, suggests that street expectations for out‐year semiconductor earnings remain too conservative. The momentum-driven build-out of the semiconductor market is expected to continue, with significant capex numbers expected to rise again in Q3; however, in the short term, investors should expect volatility driven by the unwinding of token maxing, where companies feel compelled to keep spending on the latest, most efficient models.

The trend dominated early 2026, but ultimately forced some organizations to implement monthly spending caps per employee. This dynamic mirrors a broader structural shift captured by the Jevons paradox, the economic observation that increasing the efficiency of a resource paradoxically increases its total consumption. Recent analysis from Bain & Co. captures this disconnect in real time.

According to the data, enterprise token consumption rose by 450 percent from 2024 to 2025 as unit token costs halved. As credits are rationalized and unit economics scrutinized, periodic pullbacks are likely, but as long as order books remain full and capex budgets reset higher, the structural floor for semiconductor profits appears elevated. Meanwhile, Mersch sees a clear split between struggling application-layer giants like Salesforce (NYSE:CRM), Adobe (NASDAQ:ADBE), Workday (NASDAQ:WDAY) and stronger infrastructure players like Snowflake (NYSE:SNOW), MongoDB (NASDAQ:MDB), Datadog (NASDAQ:DDOG) and DigitalOcean (NYSE:DOCN).

The latter group, he said, is beginning to show improving fundamentals, while parts of the application stack face growth deterioration, debt stress and the risk of security breaches. For investors, that suggests the next leg of AI-driven returns within software is more likely to accrue to infrastructure enablers than to mature enterprise applications still trying to reassert pricing power in a world where the marginal cost of software keeps trending toward zero. Funding and collateral in the AI space The AI infrastructure expansion is being funded by a sophisticated capital stack, with tech titans already attracting hundreds of billions of dollars in year-to-date investment.

These inflows are driving expectations that annual data center spending will hit US$750 billion by year end. While the majority of these companies maintain long-term profitability, reliance on heavy debt to fund AI expansion has faced specific scrutiny from analysts. Additionally, tech giants are still selling massive amounts of stock to finance their infrastructure, marking a potential shift in market structure from buybacks to an issuance deluge.

Hyperscalers, which have robust cash-generation abilities, are still finding ample demand for investment‐grade debt; however, more highly leveraged vendors are venturing into higher-risk segments. As Mersch explained: “There’s still a lot of confidence in their free cashflow, because they have these underlying businesses that generate a lot of free cash, whether it’s the search side from Google (NASDAQ:GOOGL), whether it’s the ad side from Meta Platforms (NASDAQ:META) or whether it’s kind of the e‐comm and cloud business from AWS and from Amazon (NASDAQ:AMZN). “Where it gets a little bit more indicative of the market is the Oracle (NYSE:ORCL) side of it, because if you look at a lot of their leverage ratios, and you look at the amount of debt they’re issuing versus their free cashflow, it is a little bit more highly leveraged and not as reliable in terms of whether they can pay that off at the end of it or not.”

Mersch cited Broadcom's (NASDAQ:AVGO) recent deal structure — asset-backed financing — as an innovative funding mechanism in a market exploring more ways to finance enormous debt tied to AI. In a recent deal, Apollo Global Management (NYSE:APO) and Blackstone (NYSE:BX) agreed to a US$10 billion loan secured by Broadcom’s custom TPUs, with the chipmaker guaranteeing a residual value on the hardware. For now, this expanding toolkit suggests that the market has multiple avenues to keep financing AI infrastructure, rather than a single point of failure; however, the interconnectivity of these financing relationships carries the risk of amplifying any future stress in AI unit economics into broader funding strains.

Investors watch next AI IPOs Following SpaceX's (NASDAQ:SPXC) blockbuster US$85.7 billion debut, which took place in mid-June, investors' attention is increasingly shifting toward the next phase of the AI cycle. While Mersch sees SpaceX as structurally important, with a credible near‐term AI/cloud revenue engine and long‐dated optionality in orbital data centers and connectivity, he noted that it's trading on a very rich valuation. “When they went out with the initial public offering (IPO), that was only 4 percent float.

Now, the Nasdaq-100 (has) accelerated their inclusion to 15 days, so it’s going to be entering that index, which is forced buying," he said. "At the same time, there’s going to be stock that comes off lockup at 70, 90, 120 and 150 days ... a massive cliff is coming six months after it goes public, in terms of a flood of supply coming into the market," Mersch added.

"So it’s going to be a very volatile ride on its way there.” In his opinion, the more transformative market moment may come when OpenAI and Anthropic eventually list. “I think both these companies are going to be issuing stock significantly,” he said, noting Anthropic’s explosive annual recurring revenue growth and predicting that both it and OpenAI will manage dilution with flexible IPO price ranges.

“Those IPOs are going to be extremely hot, because there’s been nowhere that you can invest in the application market in the public markets right now. AI has entirely been a physical economy semiconductor infrastructure play. This is the first time that it’s going to be the application layer," Mersch continued.

"A lot of these technology waves, where the relative value accrues over the longer term, typically are the application layer. This is going to be the first pure‐play way to do that, and I think it’s going to be a super hot IPO.” With record IPO activity expected this year — Goldman Sachs (NYSE:GS) analysts expect companies to raise around US$225 billion — market concentration remains a defining feature.

Mega deals are absorbing the majority of liquidity and risk appetite, leaving smaller and mid-sized IPOs struggling to gain traction. Canada’s IPO environment, in particular is facing persistent structural challenges. While recent funding commitments and company milestones point to renewed momentum in Canada’s AI and quantum sectors, alongside a growing policy focus on talent, commercialization and data sovereignty, limited access to late-stage capital and a comparatively shallow venture ecosystem continue to constrain the path to scale.

Macro factors, specifically the US‐Iran war and persistent inflation, will continue to influence markets. But as Mersch put it, “The semiconductor and AI trade was one of those things that just kind of went up regardless of the macro sort of narrative around this.” Ultimately, he expects that supply and demand will adjust to high prices and margins, even if some cyclicality remains.

Don’t forget to follow us @INN_Technology for real-time news updates! Securities Disclosure: I, Meagen Seatter, hold no direct investment interest in any company mentioned in this article. Editorial Disclosure: The Investing News Network does not guarantee the accuracy or thoroughness of the information reported in the interviews it conducts.

The opinions expressed in these interviews do not reflect the opinions of the Investing News Network and do not constitute investment advice. All readers are encouraged to perform their own due diligence. From Your Site Articles - AI Market 2025 Year-End Review › - How to Invest in OpenAI's ChatGPT › - AI Stocks: 9 Biggest Companies › - Is Now a Good Time to Invest in AI?

› - How to Invest in Artificial Intelligence ›