ON.energy Debuts Grid-Interactive UPS for AI Data Center Demands

Powering the AI Revolution

As artificial intelligence workloads push data center power consumption to unprecedented levels, ON.energy has entered the market with what it claims is the first purpose-built UPS system specifically engineered for AI infrastructure. The Miami-based company’s medium-voltage uninterruptible power supply, dubbed AI UPS, is already being deployed at several hyperscale data centers across the United States, according to industry reports.

Powering the AI Revolution
Grid-Interactive Architecture
Competitive Landscape Heats Up

Traditional UPS systems apparently struggle with the rapid load changes and voltage fluctuations characteristic of AI computing, particularly from large GPU clusters that can draw massive power in unpredictable bursts. Sources familiar with the technology suggest ON.energy’s solution acts as a dynamic interface between data centers and the electrical grid, providing what the company describes as “grid-safe resilience” while potentially accelerating interconnection timelines that have become a major bottleneck for AI campus development.

Grid-Interactive Architecture

What sets this system apart, analysts note, is its grid-interactive capability—a significant departure from conventional UPS units that typically operate in isolation. This architecture reportedly enables data center operators to participate in energy markets, with ON.energy claiming operators could generate “millions of dollars annually per 100MW of IT load” through grid services or energy savings programs.

“We engineered AI UPS to set a clear technical standard for what grid-safe data center interconnection looks like,” Ricardo de Azevedo, ON.energy’s co-founder and CTO, stated in the company’s announcement. The system is designed to scale from megawatt to gigawatt levels in both grid-connected and off-grid environments, absorbing transient loads from GPU clusters while helping stabilize the broader electrical network.

This capability comes at a critical moment. AI data centers have emerged among the world’s largest single power consumers, with some individual campuses now requiring as much electricity as medium-sized cities. The ability to manage these massive, fluctuating loads while maintaining grid stability has become a central challenge for utilities and developers alike.

Competitive Landscape Heats Up

ON.energy isn’t alone in recognizing this emerging market segment. Earlier this month, Solidion Technology announced its Peak Series UPS, which the company said specifically targets AI data center developers. This convergence of energy storage and data center infrastructure reflects how the unique power requirements of artificial intelligence are reshaping multiple technology sectors simultaneously.

Founded in 2016, ON.energy brings utility-scale energy storage experience to the table, with more than 400MWh of projects currently under construction. This background in large-scale energy management appears to be informing their approach to the data center market, where power reliability and quality have become non-negotiable requirements for AI operations.

As AI continues its rapid expansion, the infrastructure supporting it is evolving just as quickly. The emergence of specialized power systems like ON.energy’s AI UPS suggests we’re entering a new phase of data center design—one where power management isn’t just about backup, but about active participation in the broader energy ecosystem.

References

This article aggregates information from publicly available sources. All trademarks and copyrights belong to their respective owners.

Note: Featured image is for illustrative purposes only and does not represent any specific product, service, or entity mentioned in this article.

Researchers have determined the atomic structure of human glucose-6-phosphate transporter using cryo-electron microscopy. The findings reveal how the transporter recognizes its substrate and how inhibitors block its function, providing crucial insights for treating glycogen storage disease type 1b.

Atomic Structure of Glucose Transporter Revealed

Scientists have successfully determined the atomic-level structure of the human glucose-6-phosphate transporter (G6PT) using cryo-electron microscopy, according to research published in Nature Communications. The breakthrough study provides the first detailed look at how this crucial metabolic transporter functions and how mutations cause glycogen storage disease type 1b (GSD-1b), potentially opening new avenues for therapeutic development.