Changpeng Wan
The artificial intelligence revolution has reached a critical inflection point. Nvidia’s remarkable nine-quarter streak of beating expectations is showing signs of fatigue, while technology giants face mounting pressure to transform AI promises into tangible financial results. Supply chain constraints, infrastructure bottlenecks, and early signals of growth deceleration are testing investor confidence in the AI boom that has captivated markets for nearly two years.
The Diminishing Wow Factor in AI Earnings
Nvidia’s latest quarterly performance, while still exceeding analyst expectations, represents the smallest beat since the AI frenzy propelled the chipmaker to unprecedented heights. The company’s shares dropped approximately 6% following the earnings release, signaling that investors are becoming more discerning about AI progress.
From Spectacular to Sustainable Growth
Nvidia’s annual revenue growth slowed to 56% in the latest quarter, marking the slowest pace in over two years. This normalization was inevitable given the laws of large numbers—the bigger a company becomes, the more difficult it is to maintain explosive growth rates. If analyst projections hold true, the current quarter will show further deceleration. The days of stunning 53% revenue beats above expectations, like the company delivered in May 2022, appear to be fading into memory.
The Supply Chain Reality Check
UBS analysts highlighted that Nvidia’s ‘supply’ metric—the sum of inventory and supply commitments—climbed from approximately $41 billion last quarter to about $45 billion. This data suggests that while supplies remain adequate to support reasonable revenue growth, they don’t预示 the dramatic ‘hockey stick’ growth trajectory that investors had come to expect.
Structural Challenges Constraining AI Expansion
Even as demand for AI capabilities continues to grow, two significant structural challenges are emerging that could limit how quickly technology companies can deliver substantive evidence of AI progress.
The Complex Supply Chain Web
Nvidia’s AI chips are almost exclusively manufactured by TSMC, then packaged into complex configurations and inserted into increasingly sophisticated computing systems. CEO Jensen Huang (黄仁勋) revealed in March that the company’s largest AI computing systems contain approximately 600,000 components, with future systems expected to incorporate 2.5 million parts. Huang noted in a June company meeting that while supply for current systems ‘isn’t particularly difficult to obtain’ if demand can be predicted to suppliers, challenges remain. The timeline from chip manufacturing to AI supercomputer delivery to customers currently stretches to about one year.
The Power Infrastructure Bottleneck
Perhaps the more concerning limitation is emerging beyond the semiconductor supply chain: global power grids are not expanding rapidly enough to meet AI computing demands. U.S. utility companies are hesitant to develop energy infrastructure for large AI projects because they’re uncertain whether the AI boom will last long enough to justify recouping their massive investments. This electricity infrastructure gap represents a fundamental constraint on how quickly AI capabilities can scale.
Tech Giants Under Microscope for AI Delivery
With Nvidia’s growth moderating, attention is shifting to other technology giants who must now demonstrate they can convert AI investments into measurable business outcomes.
Apple’s AI Crossroads
Apple faces particularly intense pressure to showcase substantive evidence of AI progress during its product events. Despite adding $430 billion in market capitalization since late July, investors are questioning whether the company’s perceived lagging position in artificial intelligence justifies its elevated valuation. Clayton Allison of Prime Capital Financial expressed the prevailing skepticism: ‘It’s difficult to recommend establishing or adding to positions ahead of this event, especially after this run-up, because we’re not expecting to see truly exciting features that would make you want to buy. If Apple continues to stumble on AI, I worry about the stock performance.’ David Katz of Matrix Asset Advisors added: ‘The biggest risk has eased, but until we have a clearer AI roadmap, there won’t be another leg up. I think the stock goes higher long-term, but near-term there won’t be a lot of appreciation.’
Oracle’s Ascent in AI Infrastructure
Oracle is traveling a different path than Apple. After years of sluggish revenue growth, the company is benefiting from the AI compute race. Its stock has surged more than 40% this year, placing it among the top 30 performers in the S&P 500. This growth is driven primarily by its cloud infrastructure business, which Oracle expects will see sales growth leap to over 70% in the current fiscal year. Paul Meeks, Managing Director at Freedom Capital Markets, notes: ‘What I’m focused on in this earnings report is whether cloud growth can be sustained. Oracle has proven its cloud business is benefiting from AI, but now it needs to confirm that again.’
The Path Forward: From Promise to Performance
The technology sector stands at a pivotal moment where visionary AI promises must transform into demonstrable business results. Investors who once rewarded potential are now demanding substantive evidence of AI progress that translates to revenue growth, margin expansion, and sustainable competitive advantages. Companies that cannot provide this evidence risk significant valuation corrections. The coming quarters will separate AI leaders from laggards based on one critical criterion: the ability to deliver concrete results rather than aspirational projections. For investors, the message is clear—look beyond the hype and demand measurable outcomes. Scrutinize capital allocation toward AI initiatives, evaluate implementation timelines, and assess whether promised efficiencies and capabilities are materializing in financial statements. The era of giving technology companies credit for AI potential is giving way to a new phase requiring proof of execution.
