From AI adoption to monetisation

The recent tech sell offs have highlighted growing market concerns over the sustainability of the AI cycle. Rising funding costs, high valuations, intensifying competition and earnings uncertainty have dented investor sentiment and put returns on AI infrastructure spending much more firmly into focus. Nonetheless in our assessment, demand for AI infrastructure remains robust. At the same time, AI adoption is also broadening beyond model training and infrastructure build-out toward enterprise and consumer applications, particularly Agentic AI.

As enterprises increasingly deploy Agentic AI to grow revenue, enhance operational efficiency and improve profitability, and consumer AI applications continue to expand their user bases, we believe the commercialisation of AI remains firmly on track.

Many companies have adopted AI tools; according to Microsoft’s AI Research Institute, global AI adoption is rising. But adoption is only the starting point. The key challenge is deployment i.e. embedding AI into products, workflows and business processes at scale.

This distinction matters because monetisation is driven not by access to AI, but by its ability to generate measurable outcomes. Investors are increasingly looking for evidence that companies can convert AI usage into productivity gains, cost savings and new revenue streams. Deployment has become the critical bridge between adoption and monetisation.

Asia appears relatively well positioned as the AI cycle moves from model development towards broader deployment, although the extent to which these advantages translate into sustainable earnings will vary considerably across companies and industries.

Still Asia has some structural advantages:

First, the region sits at the heart of the global technology manufacturing ecosystem. Its semiconductor, memory, advanced packaging, server, networking and power infrastructure companies can benefit from rising AI investment before monetisation becomes visible at the software layer.

Second, Asia offers large and varied application environments. Its extensive digital platforms and integrated ecosystems can support faster deployment across areas such as commerce, financial services, gaming, customer engagement and workflow automation.

Third, Asia is strongly positioned in robotics, industrial automation and physical AI. As AI moves beyond digital applications into factories, vehicles and other physical environments, the region’s manufacturing depth could become increasingly important.

According to the Digital Realty 2026 Global Data Insights Survey, which surveyed 707 IT decision-makers across Australia, Hong Kong, India, Indonesia, Japan, Singapore, and South Korea, approximately 36% of respondents in Asia Pacific reported that AI has already generated measurable business outcomes.

There is growing evidence that AI is generating tangible returns across Asia’s technology sector, although the level of impact varies by industry and position within the value chain. Some of the clearest evidence of AI-related earnings has so far emerged among infrastructure and hardware suppliers, where strong investment has supported volumes, pricing and earnings.

The opportunity is also beginning to broaden. Software companies are using AI to improve productivity, shorten development cycles and automate business processes, while cloud platforms are increasingly commercialising AI services for enterprise customers.

In China, Alibaba’s Qwen large language model is being commercialised through the Bailian platform, enabling enterprise customers to deploy AI across customer service, marketing, content generation, and workflow automation. This represents one of the most tangible examples of AI monetisation through enterprise AI services and software platforms in Asia.

This suggests AI is starting to create value not only through hardware demand but also through operational efficiency and new revenue opportunities.

Agentic AI could represent an important test of whether rising AI usage can translate into scaled monetisation.

At present, Agentic AI deployment remains primarily enterprise-focused, with companies using it to automate multi-step workflows, process data and execute tool calls to improve operating efficiency and reduce costs.

As AI agents evolve from simply answering questions to executing tasks, AI compute architecture is likely to become more balanced across central processing units (CPUs), graphics processing units (GPUs), memory, storage and networking.

For the Asian technology supply chain, this suggests that if deployment scales successfully, potential beneficiaries could broaden beyond the current GPU-led ecosystem to other parts of the server supply chain, including CPUs, integrated circuit substrates, dynamic random-access memory, NAND/SSDs, networking, optical communications and power supplies.

Fig 1: 2026 Memory market growth estimates

Source: Morgan Stanley, Feb 2026

Looking further ahead, the transition from digital AI to physical AI could create new opportunities across robotics, automation and autonomous driving.

Despite the enthusiasm surrounding AI, widespread deployment does not guarantee sustainable earnings growth. Many companies are seeing growth in AI usage, but far fewer have demonstrated that customers are willing to pay more for AI-enabled products and services.

Competition could create further pressure. The rise of increasingly capable open-source models is lowering barriers to entry and accelerating the commoditisation of some AI capabilities. As AI becomes more widely available, maintaining differentiation and capturing economic value may become more challenging.

Security, governance and model reliability also remain important considerations. As AI systems become more deeply embedded in business processes and enterprise workflows, companies will need to address concerns around data security, regulatory compliance and model accuracy before deployment can scale further. In addition, rollout of AI data centres could also run into some legal and environmental hurdles.

While demand for key AI components remains strong, attention is increasingly shifting from supply constraints towards the cost and availability of capital. Higher funding costs could raise investment hurdles and place greater emphasis on capital discipline and the economics of AI infrastructure investment.

The AI investment cycle is therefore entering a more demanding phase. Opportunities continue to emerge across Asia’s technology ecosystem, but investors are likely to become increasingly selective, focusing on companies that can translate AI adoption and deployment into sustainable earnings growth and attractive returns on capital.