Asian stocks deliver mixed performance as Nvidia leadership turmoil and Anthropic IPO speculation unfold

Gains in Japan’s Nikkei 225 and China’s Shanghai Composite contrast with losses in Hong Kong’s Hang Seng and South Korea’s KOSPI.
Ahead of the Trump-Xi summit, the resignation of Nvidia CEO Jensen Huang fuels concerns that the US could tighten chip export restrictions.
Six times oversubscribed already, Anthropic’s forthcoming IPO draws strong investor demand and lifts Japanese technology shares.

A 0.91% gain took Japan’s Nikkei 225 Index to around 61,370 on Friday. Asian equities were mixed: China’s Shanghai Composite advanced 0.22% to 3,660, but Hong Kong’s Hang Seng Index lost 0.82% to 23,810 and South Korea’s KOSPI slipped 0.49% to near 5,320.

TMGM Analysis: Financial Market News, Economic Calendar & Market Insights

A broader rebound in regional semiconductor stocks failed to support Hong Kong’s Hang Seng Index. Sentiment suffered as investors weighed Nvidia’s leadership turmoil and the possible repercussions for China’s technology sector. Following reports of extramarital affairs, the scandal led Nvidia CEO Jensen Huang to step down, with CFO Colette Kress taking over as interim CEO. The episode has rekindled concerns about tighter US export restrictions and disruptions to advanced chip supplies for Chinese AI developers.

Caution ahead of next week’s Shenzhen summit between United States (US) President Donald Trump and Chinese President Xi Jinping left mainland Chinese equities divided. The Shanghai Composite gained, whereas the Shenzhen Component slipped. Allegations concerning Huang’s ties to a suspected Chinese spy have added uncertainty to the high-stakes meeting. They threaten to revive bilateral tensions over semiconductor trade and could give US lawmakers further grounds to press for tougher technology controls.

Earlier gains in South Korea’s benchmark KOSPI evaporated, while the technology-heavy Kosdaq finished higher. Trading in major AI beneficiaries was subdued. SK Hynix was unchanged, and Samsung Electronics posted a modest gain after reports that it is negotiating to buy US-listed data centre cooling provider Modine Manufacturing to broaden its AI infrastructure presence.

In Japan, a reassessment of the wider implications of Nvidia’s leadership changes for artificial intelligence helped lift the Nikkei 225 Index. Technology and AI-linked shares drove the recovery, including Kioxia Holdings, Advantest, and Tokyo Electron. Enthusiasm for Anthropic’s IPO next week provided support, with orders already running at roughly six times the shares available.

China chip trade faces deeper risks from rising geoeconomic tensions

The consequences of the Nvidia saga may reach beyond equity market volatility and into wider geoeconomic relations, Rabobank warns. Its analysts caution that “even worse, if the US decision to look the other way on chip sales was a geoeconomic bribe to ensure Chinese rare earths and help in geopolitics, then things could get far more turbulent,” highlighting how a review of export-control enforcement could destabilise already vulnerable supply chains and dependencies on strategic resources.

AI stocks: key questions answered

Artificial intelligence: the core concepts

At its core, artificial intelligence is an academic field devoted to replicating human cognitive functions, logical understanding, perception and pattern recognition in machines. Commonly shortened to AI, it encompasses artificial neural networks, machine learning or predictive analytics, symbolic reasoning, deep learning, natural language processing, speech recognition, image recognition and expert systems. The field’s ultimate objective is artificial general intelligence, or AGI: a machine capable of solving arbitrary problems without having been trained to solve them.

AI applications and their uses

Artificial intelligence serves a wide range of purposes. Among the best-known applications are generative AI platforms, including ChatGPT and Google’s Bard platform, which draw on training with large language models (LLMs) to respond to text-based queries. For original images generated from text supplied by users, Midjourney is one example. Other AI systems apply probabilistic methods to assess an entity’s attributes or how it is perceived. Upstart’s lending platform, for instance, uses an AI-enhanced credit rating system to evaluate applicants’ creditworthiness, searching the internet for information about their careers, wealth profiles and relationships. In pharmaceutical research, other forms of AI draw on large databases of scientific studies to propose potential new drugs for laboratory testing. Content aggregators such as YouTube, Spotify and Facebook collect and organise information on viewing habits, using AI applications to recommend personalised content.

Stocks with exposure to AI

Semiconductor company Nvidia (NVDA) produces AI-focused computer chips as well as some of the platforms engineers use to develop AI applications. Because it supplies the tools for further applications of artificial intelligence, many supporters regard Nvidia as a picks-and-shovels investment in the AI revolution. Another company in the sector is Palantir Technologies (PLTR), a “big data” analytics business with substantial US intelligence community contracts. Those agencies use its Gotham platform to analyse data, identify intelligence leads and support pattern recognition. Large corporations use Palantir’s Foundry product to monitor employee and customer data for predictive analytics and anomaly detection. Microsoft (MSFT), meanwhile, owns a substantial stake in OpenAI, the privately held creator of ChatGPT. OpenAI’s technology has been incorporated into Microsoft’s Bing search engine.

AI and the risk of another stock market bubble

More than 200% was Nvidia’s share price gain in the six months after ChatGPT’s public launch in late 2022. Many other AI-linked stocks also rallied following the release, prompting immediate debate among Wall Street commentators about another technology bubble. Renowned investor Stanley Druckenmiller, who has held sizeable positions in both Palantir and Nvidia, argued that bubbles never last only six months. If enthusiasm for AI did turn into a bubble, he said, extreme valuations would persist for at least two and a half years or longer, as with the DotCom bubble in the late 1990s. By the midpoint of 2023, the best assessment was that the market was not in a bubble, at least for the time being. Nvidia was indeed valued at 27 times forward sales then, but analysts expected exceptionally strong revenue growth for years ahead. For comparison, the NASDAQ 100 traded at 60 times earnings at the DotCom bubble’s peak, against 25 times earnings in mid-2023.