Goldman Sachs trader Callahan pointed out that the focus of technology stocks in 2026 will shift from hardware hype to AI return on investment and market breadth. Although the 'Mag7' still contributes significantly, growth is slowing and divergence is intensifying. He listed ten key questions, covering the sustainability of AI spending, software valuation recovery, Apple's positioning, the landscape of LLMs, and cyclical turning points, among others.
After three consecutive years of strong returns in technology stocks but accompanied by significant volatility, as 2026 approaches, Wall Street's focus is shifting from pure hardware hype to a deeper examination of artificial intelligenceReturn on Investmentand the sustainability of market breadth.
According to the latest outlook report released by Peter Callahan, Goldman Sachs' top technology trader, although the Nasdaq 100 Index will ultimately rise by more than 20% in 2025, it will not be an "easy win" year. Callahan pointed out that while the "Magnificent Seven" collectively contributed approximately $3.5 trillion in market capitalization growth in 2025, this growth rate has already shown signs of slowing compared to $5.4 trillion in 2024 and $4.8 trillion in 2023. Additionally, internal market divergence has reached an extremely high level.$NASDAQ 100 Index (.NDX.US)$More than 30% of the components ended the year in negative territory.

As investors increasingly focus on whether generative AI (GenAI) can deliver on its hefty capital expenditure promises over the next 12 months, market sentiment is undergoing subtle shifts. Callahan emphasized that the core of the current debate lies in the trajectory of AI infrastructure spending — for instance, NVIDIA data suggests annual spending could reach $3 to $4 trillion by 2030 — and whether such massive investments can translate into tangible productivity gains.
To clarify this complex market environment, Callahan outlined ten key questions that will determine the trajectory of tech stocks in 2026. These questions not only pertain to specific sector rotations but also touch upon the fundamental logic of macroeconomic and technological cycles.
The Ten Key Questions That Will Determine the 2026 Trend
Callahan explicitly raised the following ten questions in his report, which will dominate market narratives in 2026:
Where will the AI debate lead? Will the focus shift to "physical AI" (robots, autonomous vehicles, smart glasses)? Which companies will emerge as winners in productivity enhancement? How will regulation and return on invested capital (ROIC) evolve?
How will (application) software companies repair their valuations? What are the potential risks facing the software industry in the next 12-24 months? Will it be the end of seat-based pricing models, the rise of agents, usage issues, or commoditization pressures brought by large language models (LLMs)?
What is Apple's storyline? Entering 2026,$Apple (AAPL.US)$Is it a defensive growth stock or part of the AI narrative? Can foldable smartphones become a catalyst? Why is App Store growth slowing?
What are the broader impacts of the commodities supercycle? Considering price trends in memory products like DRAM, HDD, NAND, as well as gold, silver, and copper, where else is supply tight? Who can absorb rising prices, and who cannot?
What does the GenAI-driven 'efficiency' imply? If it means layoffs, will the market view it as a positive sign of productivity improvement or as a negative factor pressuring the economy and non-farm employment data?
In the debate over profit margins and competition, which internet companies are the most worthy of investment? For instance, investors are fiercely debating the prospects of companies like META.
Has the turning point for cyclical industries arrived? Will 2026 witness a cyclical reversal in housing, commercial real estate (CRE), ISM data below 50 for three consecutive years, analog chips, or the automotive industry?
Can hardware and semiconductor AI stocks lead the market again? Or will debates over gross margin, spending visibility, or intensifying competition suppress market sentiment?
How will the market's perception of large language models (LLMs) evolve? Will they move toward 'commoditization'? Will it be a competitive market with multiple participants or one dominated by a few players? Will it lean toward artificial general intelligence (AGI) or artificial superintelligence (ASI)? What role will Chinese models play? Will they advance toward productized implementation or remain in the realm of 'primitive intelligence' comparisons?
What are the blind spots now? What is currently unmentioned but will become a consensus topic by 2026? Could it be agentic commerce, the resurgence of SaaS stocks, or specific use cases of AI productivity?
2026 Outlook: Searching for Second Derivatives and Mean Reversion
Looking back at 2025, the most prominent feature of the market was 'differentiation.' Callahan pointed out that despite low volatility at the index level, individual stock volatility was extremely high. While technology stocks performed well overall, semiconductors and network infrastructure sectors led by an absolute advantage and were seen as the most crowded trades among investors; in contrast, telecommunications, payments, and application software sectors lagged behind.
Looking ahead to 2026, Callahan believes that the return prospects for the Nasdaq 100 Index remain robust, but gains may be more concentrated in the first half of the year. This is because the index has recently undergone a period of consolidation and faces a 'wall of worries' such as the sustainability of AI spending, which often favors stock market rallies due to lowered expectations.
In terms of investment themes, Callahan recommends focusing on the market's 'broadening' trade, where capital flows from crowded AI infrastructure stocks into other areas. He believes that in 2026, investors will seek the 'second derivatives' of AI—discounted stocks that leverage AI to reduce costs, improve product discovery, or drive new revenue streams—not just the 'shovel sellers' like hardware suppliers.
Editor/melody