|
China is emerging as the strongest long-term contender in the artificial intelligence (AI) race despite ongoing volatility in global semiconductor stocks, according to a report by Jefferies.
The report said recent sharp declines in semiconductor stocks have brought many of them close to their 200-day moving averages, raising questions over whether the correction is merely a technical unwinding of leveraged positions or an early signal of slowing capital expenditure by global hyperscalers. Despite the market volatility, Jefferies maintained that China remains well positioned in the AI race, particularly in consumer-facing applications. "The Chinese are best positioned to prevail in AI, most particularly in the mass consumer market," the report said. Jefferies noted that while concerns have increased over massive AI-related spending, the latest quarterly earnings have not yet indicated any broad-based slowdown in AI investment. As a result, analysts have not reduced their earnings forecasts for memory-chip companies. However, it added that markets have started reacting negatively to rising capital expenditure, calling it an important warning sign. The report highlighted differing investor reactions to recent earnings announcements by major technology companies. Alphabet came under pressure after reporting negative free cash flow in the second quarter of 2026 for the first time since its initial public offering in 2004. Similarly, Meta's shares declined 10 per cent in after-hours trading after its free cash flow plunged 91 per cent, falling from USD 8.5 billion in the second quarter of 2025 to USD 784 million in the second quarter of 2026. At the same time, the company raised its 2026 capital expenditure guidance to USD 130-145 billion, compared with its earlier guidance of USD 125-145 billion. In contrast, Microsoft gained 8 per cent in after-hours trading after maintaining its capital expenditure guidance for calendar year 2026 at approximately USD 175 billion. The report noted that this figure was revised from the earlier guidance of USD 190 billion due to accounting changes related to the useful life of assets and the shift of finance leases to operating leases, which are not included in capital expenditure. The report further argued that while hyperscalers are investing aggressively in AI infrastructure, demand for computing power is expected to continue expanding in the AI era, although the customer base driving that demand may evolve over time. It added that AI could ultimately resemble the airline industry rather than the winner-takes-all model that characterised the internet era, suggesting that heavy capital expenditure may not necessarily translate into outsized returns for all participants. It stated, "the demand for compute will keep growing in the AI era though the customers may change". (ANI)
|
