Asia's AI Strategy: STT GDC Survey Reveals Stagnation in Growth Due to Infrastructure and Talent Shortages
A survey by ST Telemedia Global Data Centres (STT GDC) across over 600 companies in 9 Asian countries indicates high AI adoption intent, yet many are stuck in the "building" phase due to infrastructure and talent deficits. This highlights challenges in achieving ROI from AI investments, low sustainability awareness, and a discrepancy in infrastructure selection criteria.
📋 Article Processing Timeline
- 📰 Published: May 11, 2026 at 20:00
- 🔍 Collected: May 11, 2026 at 11:31
- 🤖 AI Analyzed: May 11, 2026 at 12:13 (41 min after Collected)
Singapore, April 15, 2026 — ST Telemedia Global Data Centres (hereinafter, STT GDC), a Singapore-headquartered data center colocation provider, has released its latest Asia regional survey report, "The Gap Between Strategy and Reality: Infrastructure Challenges Affecting AI Strategy." This survey analyzes how companies across Asia are transitioning from the initial stages of AI utilization to the implementation phase. The survey was conducted by STT GDC in collaboration with its research partner, Ecosystem, targeting over 600 companies in nine Asian markets: India, Indonesia, Japan, Malaysia, Philippines, Singapore, South Korea, Thailand, and Vietnam.
**Despite High AI Adoption, Insufficient Readiness for Execution Phase Revealed**
According to the survey, interest and willingness to adopt AI are very high across Asia, with approximately 90% of companies already engaged in AI utilization. However, 71% of these companies remain in the "building" stage of AI maturity. In this stage, the direction of AI strategy is established, and initial initiatives such as Proof-of-Concept (PoC) are underway, but they have not yet progressed to continuous utilization in a production environment. As a result, many companies are struggling to generate stable and measurable Return on Investment (ROI).
**Challenges Faced During the AI Building Phase**
In a situation where demonstrating measurable ROI is difficult, this creates a challenge in justifying additional investment in high-density, purpose-built infrastructure environments. Furthermore, the lack of in-house specialized talent exacerbates the challenges in the building phase. The reality is that many companies lack sufficient specialized operational skills and knowledge required to operate and manage increasingly complex and sophisticated AI infrastructure at scale.
Chris Street, Group Chief Revenue Officer at STT GDC, commented: "Across Asia, companies are rapidly moving from AI proof-of-concept to implementation. However, the success of AI is no longer solely dependent on the development of AI models themselves, but rather on the establishment of the supporting infrastructure and operational readiness. Without scalable infrastructure and adequate operational preparedness, it will be difficult to continuously convert the initial expectations of AI into stable enterprise value."
**Sustainability Challenges in AI Expansion**
Despite the rapid increase in power consumption and cooling demands accompanying the expansion of AI workloads, it became clear that sustainability considerations are still often neglected in infrastructure selection for many companies. Only 27% of companies responded that ESG goals would be an important or central factor in their future planning. Meanwhile, 64% of companies across Asia still prioritize performance and cost. Although power density, thermal efficiency, and long-term Total Cost of Ownership (TCO) are becoming crucial for responsibly scaling AI, many companies' awareness has not yet fully caught up.
**The Gap Between Corporate "Perception" and "Reality" Hinders AI Adoption**
The survey also revealed a significant gap between the capabilities companies actually need to fully scale AI adoption and the criteria they use to evaluate infrastructure partners. Across Asia, many companies cite operational expertise, scalability, and cost efficiency as key challenges in AI utilization. However, when selecting infrastructure partners, the reality is that they still prioritize basic requirements such as security and reliability. As a result, the functions and systems essential for full-scale AI deployment are not adequately reflected in evaluation and decision-making processes.
**AI Adoption in Singapore and Future Constraints**
While these challenges are observed across Asia, they manifest differently in more mature AI markets. Singapore, in particular, stands out with a higher proportion of companies moving beyond initial PoCs compared to the regional average. While only 17% of companies across Asia are in the "integration" stage of AI, considered ready for full-scale AI deployment, 40% of companies in Singapore have already reached this stage, demonstrating higher initial execution power and adoption.
However, the survey also revealed that advancing to the next "leading" stage of AI is the most difficult challenge. Only 3% of Singaporean companies have reached the "leading" stage in AI infrastructure maturity, indicating that even in Singapore, considered Asia's most mature AI market, progress to the highest level is limited.
**Despite High AI Adoption, Insufficient Readiness for Execution Phase Revealed**
According to the survey, interest and willingness to adopt AI are very high across Asia, with approximately 90% of companies already engaged in AI utilization. However, 71% of these companies remain in the "building" stage of AI maturity. In this stage, the direction of AI strategy is established, and initial initiatives such as Proof-of-Concept (PoC) are underway, but they have not yet progressed to continuous utilization in a production environment. As a result, many companies are struggling to generate stable and measurable Return on Investment (ROI).
**Challenges Faced During the AI Building Phase**
In a situation where demonstrating measurable ROI is difficult, this creates a challenge in justifying additional investment in high-density, purpose-built infrastructure environments. Furthermore, the lack of in-house specialized talent exacerbates the challenges in the building phase. The reality is that many companies lack sufficient specialized operational skills and knowledge required to operate and manage increasingly complex and sophisticated AI infrastructure at scale.
Chris Street, Group Chief Revenue Officer at STT GDC, commented: "Across Asia, companies are rapidly moving from AI proof-of-concept to implementation. However, the success of AI is no longer solely dependent on the development of AI models themselves, but rather on the establishment of the supporting infrastructure and operational readiness. Without scalable infrastructure and adequate operational preparedness, it will be difficult to continuously convert the initial expectations of AI into stable enterprise value."
**Sustainability Challenges in AI Expansion**
Despite the rapid increase in power consumption and cooling demands accompanying the expansion of AI workloads, it became clear that sustainability considerations are still often neglected in infrastructure selection for many companies. Only 27% of companies responded that ESG goals would be an important or central factor in their future planning. Meanwhile, 64% of companies across Asia still prioritize performance and cost. Although power density, thermal efficiency, and long-term Total Cost of Ownership (TCO) are becoming crucial for responsibly scaling AI, many companies' awareness has not yet fully caught up.
**The Gap Between Corporate "Perception" and "Reality" Hinders AI Adoption**
The survey also revealed a significant gap between the capabilities companies actually need to fully scale AI adoption and the criteria they use to evaluate infrastructure partners. Across Asia, many companies cite operational expertise, scalability, and cost efficiency as key challenges in AI utilization. However, when selecting infrastructure partners, the reality is that they still prioritize basic requirements such as security and reliability. As a result, the functions and systems essential for full-scale AI deployment are not adequately reflected in evaluation and decision-making processes.
**AI Adoption in Singapore and Future Constraints**
While these challenges are observed across Asia, they manifest differently in more mature AI markets. Singapore, in particular, stands out with a higher proportion of companies moving beyond initial PoCs compared to the regional average. While only 17% of companies across Asia are in the "integration" stage of AI, considered ready for full-scale AI deployment, 40% of companies in Singapore have already reached this stage, demonstrating higher initial execution power and adoption.
However, the survey also revealed that advancing to the next "leading" stage of AI is the most difficult challenge. Only 3% of Singaporean companies have reached the "leading" stage in AI infrastructure maturity, indicating that even in Singapore, considered Asia's most mature AI market, progress to the highest level is limited.