Morgan Stanley IM Executive Sees AI Data Center-Related Investment Reaching About $5 Trillion by 2030
DataNews Editorial Team
Vishal Khanduja, head of fixed income at Morgan Stanley Investment Management, said global investment related to AI data centers could reach about $5 trillion by 2030. He also warned that at least half of industry participants may fail to generate returns above their cost of capital over the next five years.
- Morgan Stanley Investment Management's Vishal Khanduja forecasts about $5 trillion of global AI data center-related investment by 2030.
- Khanduja said at least 50% of industry participants may not achieve returns above their cost of capital over the next five years.
- The outlook frames the AI infrastructure cycle as both capital-intensive and highly uncertain on returns.
- Potential capital inflows could be substantial, but may produce sharply differentiated project and investor outcomes.
Speaking on 16 October 2026 at the Global Wealth Management Forum at the 2026 Shanghai Suhewan Conference, Vishal Khanduja of Morgan Stanley Investment Management said global investment related to AI data centers is expected to reach approximately $5 trillion by 2030.
Khanduja said the scale of investment continues to increase, while characterising the current AI investment cycle as one marked by substantial market disagreement and high uncertainty. He cautioned that, over the next five years, at least 50% of industry participants may not generate returns above their cost of capital.
For data-center investors and operators, the combination of these two views is material: a very large prospective funding cycle does not necessarily translate into uniformly viable infrastructure assets. If the forecast is realised, the sector could see a major inflow of capital into AI infrastructure; however, Khanduja’s return warning indicates that project and investor outcomes may diverge significantly. Access to capital alone may not ensure the economic viability of data-center assets.
The $5 trillion figure is a forecast of AI data center-related investment by 2030, not a measure of capital already deployed. No methodology, country or operator breakdown, data-center capacity detail, individual-project CAPEX, or investment timing profile was disclosed.
The forecast points to a potentially exceptional volume of capital flowing into AI infrastructure and data centers. But the warning on returns suggests that investment growth could outpace economically justified demand and projects’ ability to earn returns above the cost of capital.
- — / October 16, 2026 · ChineseSource →
This article was prepared by the DataNews editorial team based on the sources listed above.