Swiss Re Institute said in Monte Carlo on 5 September 2026 that the global investment supercycle in artificial intelligence data centers and energy infrastructure could generate about $200 billion in cumulative industrial insurance premiums between 2026 and 2030. The research, published in sigma 3/2026 under the title "Time to build: Expanding the frontier of insurability for the capex super-cycle," argues that the buildout is creating larger, more interconnected risks because assets are becoming more concentrated and more dependent on shared networks.
According to the institute, global energy investment is expected to reach $3.4 trillion in 2026, with about $2.2 trillion directed to renewables, nuclear power, power grids, storage, low-emission fuels, efficiency and electrification. The five largest US hyperscalers are expected to invest nearly $800 billion in AI in 2026, while global data-center investment is estimated to exceed $1 trillion.
The shift turns data centers from IT assets into strategic infrastructure. Their power demand is measured in gigawatts, their asset values in billions of dollars, and their operation depends on electricity, telecommunications, cooling systems and cloud infrastructure. Swiss Re Institute said the replacement value of a single AI data center, including IT equipment, can reach $50 billion.
Risk concentration is not limited to geography. The sigma report identifies four structural factors: larger individual assets, geographic clustering, supply-chain dependencies, and shared physical and digital networks. These factors can reinforce one another, allowing a single disruption to affect multiple insureds, sectors and business lines at once.
In the United States, Texas and Virginia account for more than 40% of existing and planned data-center capacity. More than a quarter of total US capacity is located in areas expected to experience at least three days of strong hail per year, and about 40% is in areas with at least three tornado days per year. In Asia, the report points to Taiwan, where about 88% of semiconductor manufacturing facilities are located in areas with high to very high earthquake risk. Because Taiwan is central to the global semiconductor supply chain, a major loss could also affect downstream sectors.
Supply-chain dependence adds another layer of risk. Lead times for critical components such as high-voltage transformers can be several years, potentially delaying projects and increasing business-interruption losses. Shared power and digital networks can also transmit disruptions to companies that are otherwise not directly connected.
Swiss Re Institute said the constraint is not a lack of insurance capital but whether that capital can be deployed reliably for increasingly complex risks. Many large infrastructure projects have limited operating history, making it difficult to estimate loss frequency and severity. At the same time, concentrations can produce very large losses, complicating broad risk distribution.
The operating phase is the next challenge for insurability. Construction risks are relatively well understood, but once high-value assets enter service, property, business-interruption, liability and indirect business-interruption risks increase. In some cases, financial losses from business interruption can exceed the underlying property damage.
Gianfranco Lot, Swiss Re's Chief Underwriting Officer for Property & Casualty Reinsurance, said the digital economy is becoming physical, requiring data centers, power grids and increasingly complex infrastructure that must be insured. He said capacity provision will depend on how well insurers understand and manage these risks and whether they are compensated for the associated extreme risks.
Jérôme Haegeli, Group Chief Economist and Head of Swiss Re Institute, said a new investment era is emerging, with unprecedented capital flowing into infrastructure that will drive future economic growth. He said the trend increases asset concentration and creates new dependencies on power grids, supply chains and digital networks, making insurance essential for resilience and financeability.
Swiss Re Institute said technically proficient underwriting, improved models and better management of risk concentration are needed to assess the risks reliably. Distributing exposures across insurers, reinsurers and capital markets would spread large exposures across multiple balance sheets, helping keep major infrastructure projects insurable and supporting investment that underpins future economic growth.
