The maturing AI investment cycle has led several managers to argue the next phase will reward discipline, capital efficiency and diversified exposure across semiconductors, cloud, utilities and industrial automation.
Asset allocators in Asia have highlighted artificial intelligence is shifting from its early hype cycle into a monetisation test, with companies in the region offering differentiated opportunities for investors seeking to broaden their exposure.
The maturing AI investment cycle has led several managers to argue the next phase will reward discipline, capital efficiency and diversified exposure across semiconductors, cloud services, utilities and industrial automation.
As a result, they noted investors who broaden their exposure beyond mega-cap concentration while remaining selective within the AI value-chain may be best positioned for sustainable earnings growth.
Potential SpaceX IPO a 'structural turning point' for the space economy
Cheuk Wan Fan, CIO for HSBC Private Bank and Premier Wealth in the region, emphasised that investors should use a wider lens to avoid over-concentration, with Asia a good starting point.
"As the world's technology hardware powerhouse, largest consumer and manufacturer, Asia is all set to benefit from the AI investment cycle and innovation-driven productivity gains," she said.
While North Asia accelerates growth in industrial automation, robotics and humanoid technology, Jean Chia, chief investment officer at Bank of Singapore, pointed to Southeast Asia as likely to become embedded into the semiconductor supply chain with data centre developments in the coming years.
For example, Taiwan and South Korea are attracting attention for being deeply embedded in the global AI chip supply chain, she noted, while China's stated AI ambition is to become "an intelligent economy and an intelligence society" by 2035 with AI adoption of over 90%.
Differentiating the winners
As investors eye winners in the AI monetisation phase, Chia forecast AI applications will take centre stage with agentic and physical AI broadening the investment landscape.
"We favour infrastructure buildout beneficiaries such as selected hyperscalers, internet and semiconductor players such as wafer fabrication equipment (WFE), while maintaining caution on software leaders vulnerable to disruption by AI natives rolling out commercial applications," Chia explained.
Fund managers in 'run it hot' mode with bullishness at five-year high
UBP technology investment specialist Dimitri Kallianiotis agreed the backbone of the AI boom is capital expenditure by "big tech". He highlighted projections of capital spending in 2026 to rise by more than 34%, in turn sustaining strong demand for semiconductors, networking equipment, data centre hardware and cloud infrastructure.
Valuations: wary but not worried
Even though concerns about a valuation bubble persist, some allocators explained that current evidence indicates the AI investment cycle remains in its early-to-middle stages – with returns being fuelled by tangible earnings growth rather than speculative excesses.
"Although the journey may be marked by periods of volatility, we are confident that the substantial investments in building out AI infrastructure are both justified and essential to meet the future demands of the global economy," said Kallianiotis.
He explained that the high visibility of incremental revenue from cloud computing, coupled with the significant cost-saving opportunities that AI offers to global corporations, underscores the rationale behind these investments.
"With a projected US$500bn in AI-related spending targeted for 2026, this ambitious goal appears not only achievable but also necessary to unlock the transformative potential of AI across industries," Kallianiotis added.
China doubles down on AI
Meanwhile, self-sufficiency in semiconductors and AI has become a top strategic priority, driven by geopolitical tensions, according to Vey-Sern Ling, senior equity adviser at UBP.
US restrictions on chip access have spurred rapid development in China, especially since the launch of DeepSeek's Chatbot and R1 model shocked markets in early 2025 – showcasing China's ability to produce competitive large-language models despite chip restrictions.
"Chinese AI models are high-quality, cost-efficient and open source," said Ling. "Technology companies can focus on building and monetising applications, supported by the advantage of a large domestic market."
He predicted 2026 will bring accelerating AI infrastructure spending in China, with product roll-outs and a steep ramping up of monetisation in both cloud computing and adtech.
