AI infrastructure spending is supporting rapid growth across chipmakers, memory suppliers and cloud providers. In their latest reported quarters, NVIDIA’s Data Center revenue increased 117% year over year, Broadcom’s AI semiconductor business delivered 221% growth, and Micron’s Cloud Memory unit expanded 307%. Cloud revenue growth reached 37% at AWS and 82% at Google Cloud.
However, share-price reactions were less uniform. Reported after-hours moves following their respective earnings releases included gains of 4.2% for NVIDIA, 12% for Micron and nearly 9% for Amazon, compared with declines of over 1% for Broadcom and about 3% for Google parent Alphabet.
This divergence suggests that strong growth alone is no longer enough to support further share-price gains. The focus is whether AI infrastructure leaders can continue to exceed investor expectations. A key part of that assessment is identifying where their businesses could accelerate next and how that acceleration could lift future earnings.
NVIDIA: A Platform Upgrade and Broader Adoption Support Growth
The transition from NVIDIA’s Blackwell AI computing platform to its successor, Vera Rubin, could increase revenue per infrastructure deployment. Management expects Rubin to contribute approximately 20% of Data Center revenue in Q3 FY2027 and estimates a revenue opportunity of $40 billion per gigawatt, versus $25 billion for Blackwell. Upgrades across GPUs, CPUs and networking could therefore allow NVIDIA to capture more revenue within a given power allocation, adding a growth lever alongside new data-center capacity.
Broader customer adoption could also expand sales and strengthen revenue resilience. In Q2 FY2027, revenue from NVIDIA’s ACIE group—covering neocloud, industrial and enterprise customers—grew 25% sequentially, compared with 13% in Hyperscale. Some neocloud capacity still serves hyperscalers, so the diversification benefit depends on the underlying customers. Continued enterprise and sovereign adoption would provide stronger evidence of independent demand, potentially making growth less sensitive to individual hyperscalers’ spending decisions.
Broadcom: Custom Chips and Networking Expand the Earnings Opportunity
Growing demand for custom AI accelerators and networking products could expand Broadcom’s revenue and operating-profit base. Management has explicitly highlighted these two components of its AI semiconductor business as growth drivers. Its FY2027 AI revenue forecast has risen to approximately $115 billion, from more than $100 billion, reflecting a larger opportunity as customer deployments scale.
The potential earnings benefit comes from scaling a highly profitable business, rather than requiring margins to keep rising. For Q4 FY2026, Broadcom forecasts 93% company-wide revenue growth while maintaining a 66% non-GAAP operating margin, unchanged from a year earlier. Delivering that combination would substantially expand operating profit even without a higher margin percentage—a more useful measure of the growth opportunity than revenue acceleration alone.
Amazon: AWS’s AI Expansion Could Amplify Group Earnings
Expanding AWS AI services could have an outsized effect on Amazon’s operating earnings because cloud contributes disproportionately to group profitability. At its Q2 2026 update, Amazon reported an AWS AI business exceeding a $25 billion annualized revenue run rate. Based on the company’s reported segment results, AWS accounted for approximately 21% of group revenue but 60% of operating income, making cloud growth particularly consequential for the earnings outlook.
AWS is also converting more of its growing revenue into operating profit. Revenue growth accelerated from 28% year over year in Q1 to 37% in Q2, while operating margin increased from 37.7% to 39.4%. Sustained AI adoption could therefore expand Amazon’s earnings through a larger, already-profitable business—even if AWS margins stabilize rather than continue climbing.
Alphabet: Cloud Services and TPU Sales Create Two Routes to Monetization
Cloud services and sales of TPU systems—Google’s specialized AI computing hardware—give Alphabet complementary channels for increasing Cloud’s profit contribution. Stronger demand has already prompted earnings upgrades: S&P Global’s July 27 analysis reported increases of 10.9% to Q3 Cloud revenue forecasts and 17.7% to operating-income forecasts following the latest results. The larger profit revision points to expectations for improved profitability, not simply higher sales.
Additional owned infrastructure could also improve the economics of serving that demand. Management has flagged temporary margin pressure from using higher-cost third-party computing capacity to address supply constraints. As Google brings more of its own capacity online, it could accommodate additional customers while reducing reliance on external resources. Sufficient utilization would allow that investment to support both revenue growth and more efficient delivery, after accounting for depreciation and operating costs.
Micron: HBM4 Adoption and Better Manufacturing Yields Could Support Earnings
Micron’s HBM4 ramp—a product upgrade in high-bandwidth memory for AI accelerators—could increase advanced-memory sales and strengthen its product mix. By its Q3 FY2026 earnings update, Micron had reported more than $1 billion in HBM4 revenue, with its volume ramp progressing approximately twice as fast as the comparable previous-generation HBM3E 12-high ramp. The faster rollout provides a concrete growth driver within advanced memory, rather than relying solely on broader industry price increases.
Manufacturing improvements could add another source of earnings growth. Management expects HBM4 to reach mature yields faster than the previous generation. Higher yields—the share of production that becomes usable output—could support larger deliveries and reduce ramp-related costs per saleable unit. That creates an efficiency-driven earnings opportunity alongside higher shipment volumes, helping Micron capture more value from increasingly complex memory products.
From Growth Drivers to Quality Assets
These opportunities can support further earnings upgrades, but their value depends on execution. Deployment delays can defer revenue, underused cloud capacity can weaken investment returns, and rising component costs can offset stronger sales. The next confirmation is whether demand and operational progress continue to exceed the assumptions already reflected in market forecasts.
Identifying quality assets therefore means looking beyond short-term price action to the businesses supporting future earnings. Product upgrades need to become delivered revenue; customer expansion needs to broaden underlying demand; and faster sales need to translate into sustainable profitability. Those distinctions help separate a compelling growth narrative from a stronger investment case.
That is the mindset behind Tapbit’s continued focus on quality assets: helping traders explore meaningful market opportunities through the businesses and growth drivers behind them. NVDA/USDT, AVGO/USDT, AMZN/USDT, GOOGL/USDT and MU/USDT are available to trade as contracts on Tapbit.
Visit the campaign page for eligible trading tasks, participation requirements and rewards:
Frequently Asked Questions
What are AI infrastructure stocks?
AI infrastructure stocks are companies that provide the chips, memory, networking equipment, cloud services and data-center capacity needed to develop and operate artificial intelligence systems.
What could drive NVIDIA’s next phase of growth?
NVIDIA’s growth could be supported by the transition from Blackwell to the Vera Rubin platform, higher revenue per infrastructure deployment and broader adoption among enterprise, industrial and cloud customers.
How could cloud AI growth benefit Amazon and Alphabet?
Amazon could benefit because its cloud division contributes a significant share of operating income, while Alphabet has two monetization paths through cloud services and sales of its specialized AI computing systems.
Why is HBM4 important for Micron?
HBM4 is an advanced high-bandwidth memory product designed for AI accelerators. Faster adoption and improved manufacturing yields could help Micron increase shipments, strengthen its product mix and reduce production costs.
What risks could affect AI infrastructure growth?
Deployment delays, underused cloud capacity, supply constraints and rising component costs could weaken returns. Strong demand must ultimately translate into delivered revenue and sustainable profitability.
