Nvidia reported record second-quarter revenue of $46.7 billion, a 56% rise on the year, cementing its role as the engine of the AI economy. Yet despite exceeding Wall Street forecasts, the company’s shares slipped by more than 2% in after-hours trading — a sign that investors’ nerves are not easily calmed.
This paradox — record results yet muted reaction — captures the tension surrounding Nvidia’s position at the heart of the AI boom. Its market capitalisation has surged past $4.4 trillion, making it one of the world’s most valuable companies, while its data centre division now accounts for almost 90% of total sales. The performance has left the company trading at a lofty multiple, with a trailing P/E ratio of nearly 60.
“Nvidia’s better-than-expected results underscore just how important the company has become to the AI revolution. But this technological dominance cuts both ways,” said Lukman Otunuga, senior market analyst at FXTM. “The market is effectively pricing Nvidia as a run-away leader, which leaves little margin for error. If demand softens or new competitors break through with better product, today’s lofty valuation could quickly look exposed.”
That sense of fragility is amplified by questions over how long Nvidia can hold this grip. Rivals are moving to close the gap. AMD has launched its MI350 series on a cutting-edge 3nm process, already adopted by Microsoft and OpenAI. Intel is positioning its Gaudi accelerators as lower-cost enterprise alternatives. In Europe, companies such as ASML and ARM occupy strategic points in the semiconductor supply chain, offering potential levers for market diversification.
Yet some of the most significant threats to Nvidia’s long-term dominance may come from its own customers. “Currently, Nvidia benefits from being the only chipmaker the big tech companies can turn to, but that may not be the case forever,” said George Sweeney, personal finance expert at Finder. “Meta is already spending around $65 billion in 2025 alone to create and develop its own AI chips to reduce its reliance on Nvidia. These mammoth tech companies do not want to be beholden to anyone but themselves.”
For now, Nvidia continues to ride an unprecedented wave of demand. Hyperscalers including Amazon, Microsoft, Google and Meta are set to spend more than $330 billion in 2025 on AI-focused infrastructure. Analysts project the AI chip market will expand from $85 billion last year to nearly $370 billion by 2032. But that rapid growth has triggered debate over a possible AI bubble. A recent MIT survey found 95% of businesses were not yet achieving a profitable return on their AI investments — raising questions over whether adoption is running ahead of enterprise value creation.
This divergence between massive capital flows and slower returns feeds directly into another source of unease: geopolitics. US export restrictions on high-end chips have already cost Nvidia billions, including a $4.5 billion hit linked to its H20 product line. China, which CEO Jensen Huang has described as a $50 billion opportunity, is pouring state funds into developing domestic alternatives.
“In the near term, US restrictions may slow Nvidia’s growth, but the bigger story is potential fragmentation where instead of a single global AI market, we may see competing ecosystems,” Otunuga said. “These risks are diluting efficiencies and slowing innovation, even as governments frame AI leadership as a strategic imperative.”
The chill of domination —
That backdrop helps explain why investors appeared unmoved by Nvidia’s latest $60 billion buyback programme. Normally read as a sign of confidence, the announcement fell flat. With the stock already trading at premium valuations, and with long-term demand clouded by competitive and geopolitical risks, the market is no longer rewarding gestures alone.
“Nvidia just posted another monster quarter – revenue and earnings per share exceeded expectations, while the company also announced a jaw-dropping $60 billion share buyback. That’s the kind of signal markets usually love as it says, ‘We’re confident. We’re here to stay’,” said Kate Leaman, chief market analyst at AvaTrade. “But the stock still dipped after hours – so, what caused this? The mild stock dip wasn’t about failure; rather, it was about expectations. With options traders pricing in a 6% swing, anything short of perfect was going to invite second-guessing.”
The market is demanding more than strong results. It wants certainty about Nvidia’s future direction. The company’s new Blackwell architecture is ramping at speed, with CEO Huang predicting AI-related infrastructure spending could reach $3–4 trillion globally by 2030. Nvidia is also exploring ways to adapt its products to export rules, with a successor chip designed specifically to comply with restrictions while maintaining performance. At the same time, supply chain resilience remains a priority, with closer partnerships with TSMC and new facilities in the US aimed at insulating production from geopolitical shocks.
Nvidia’s guidance suggests continued operational momentum, with a revenue forecast of $54 billion for the next quarter. Yet its trajectory depends as much on political and macroeconomic forces as on product execution.
“In the end, Nvidia didn’t disappoint. It delivered. But it also reminded the market of something important: dominance comes with pressure,” said Leaman. “When you’re leading the AI revolution, every quarter isn’t just about results, but also about keeping the entire ecosystem believing in your story.”
For investors, regulators and rivals alike, the story is no longer just about Nvidia’s quarterly results — but about how long the world’s most valuable chipmaker can sustain its lead in a market it helped to create.
