The AES Corporation

🇺🇸New York Stock Exchange
Back to all articles
The AES Corporation stock analysis AES overview
Somewhat Bullish +50

NVIDIA Stock Gains After AI Factory Deal With Major US Energy Groups

📈 NVIDIA shares rose 2.76% to $177.47 on March 23 following the announcement of a new partnership with Emerald AI and major US energy groups.

🤝 The collaboration includes AES, Constellation, Invenergy, NextEra Energy, Nscale Energy & Power, and Vistra to build AI factories with faster grid connections.

⚡ The project utilizes NVIDIA's Vera Rubin DSX AI Factory reference design and DSX Flex software library to accelerate infrastructure deployment.

🏗️ Next-generation AI factories can use co-located generation and storage as bridge power during early operations before full grid interconnection is complete.

🔄 Flexible AI factories can also connect directly to the grid using software controls to adjust power usage without needing on-site energy resources initially.

💡 Emerald AI's Conductor platform will coordinate compute flexibility with onsite generation, batteries, and behind-the-meter resources to shorten bridge power time.

🗣️ Jensen Huang stated that AI factories must be designed as a unified architecture of energy, compute, networking, and cooling.

⚠️ Varun Sivaram emphasized that valuable AI sites should support the grid rather than operating as passive loads or permanent islands.

🏭 Partner energy companies will evaluate generation applications for hybrid projects that create value for both the AI factory and the broader power grid.

💬 Constellation CEO Joe Dominguez noted that the industry faces a peak demand problem rather than a fundamental supply shortage.

🔄 AES CEO Andrés Gluski highlighted that flexible AI infrastructure can operate as a valuable asset for the electrical grid.

⚙️ NextEra Energy CEO John Ketchum stressed the need for technologies enabling quicker and lower-cost grid integration to meet new demand.

🧪 NVIDIA and Emerald AI tested AI power flexibility at five commercial data centers globally over the last year.

📍 The DSX Flex software is expected to deploy at commercial scale later this year at the NVIDIA AI Factory Research Center in Virginia.

🚀 The initiative aims to identify and advance more projects using the Vera Rubin DSX design to accelerate large-scale AI infrastructure deployment.

🌎 The partnership seeks to expand economic activity tied to AI investment and strengthen US energy capacity over time.

Bullish Signals
  • NVIDIA shares rose 2.76% to $177.47 on March 23 following the announcement of a new partnership with Emerald AI and major energy firms.
  • The strategic alliance includes top-tier entities such as AES, Constellation, Invenergy, NextEra Energy, Nscale Energy & Power, and Vistra.
  • NVIDIA's Vera Rubin DSX AI Factory reference design is positioned to accelerate grid interconnection, enabling AI sites to come online sooner.
  • Emerald AI's Conductor platform will coordinate compute flexibility with onsite generation, batteries, and behind-the-meter resources to reduce the need for oversized infrastructure.
  • CEO Varun Sivaram of Emerald AI highlighted that these sites can produce AI tokens while offering support back to the grid, creating a dual-value model.
  • AES CEO Andrés Gluski noted that flexible AI infrastructure can operate as a valuable grid asset rather than just a passive load.
  • NextEra Energy CEO John Ketchum emphasized that new demand requires technologies for quicker and lower-cost grid integration.
  • NVIDIA and Emerald AI have already tested this power flexibility at five commercial data centers globally over the last year.
  • Commercial rollout of DSX Flex is planned for later this year at the NVIDIA AI Factory Research Center in Virginia, with one of the first power-flexible factories scheduled there.
Risk Factors
  • NVIDIA stock rose only 2.76% to $177.47 on March 23, suggesting the market may not fully value the long-term impact of this partnership or is already pricing in high expectations.
  • The DSX Flex software library and architectural designs are being tested on a limited basis at five commercial data centers, with only one location planned for full-scale deployment later this year.
  • The rollout timeline is delayed until later in the year for commercial scale, indicating that widespread adoption of this flexible power model will not happen immediately.
  • The initiative focuses heavily on 'bridge power' and co-located generation as a workaround to speed up grid connections, implying underlying infrastructure constraints are already critical.
  • CEO Constellation Joe Dominguez admitted energy companies face a 'peak problem' rather than a supply deficit, highlighting structural challenges in scaling AI data center power.
  • The reliance on software controls like the Conductor platform to manage power use suggests that hardware-based solutions may be insufficient to meet rising electricity demand alone.
Full Analysis
NVIDIA shares rose approximately 2.76% to $177.47 on March 23 following an announcement of a strategic partnership focused on building next-generation AI factories capable of faster grid connections in the United States. This initiative involves a collaboration between NVIDIA, Emerald AI, and several major US energy firms including AES, Constellation, Invenergy, NextEra Energy, Nscale Energy & Power, and Vistra. The core of this partnership revolves around deploying NVIDIA’s Vera Rubin DSX AI Factory reference design alongside its new DSX Flex software library, which is engineered to optimize power usage for high-demand AI infrastructure. The agreement highlights a shift toward flexible AI factories that can utilize co-located generation and storage systems to operate as "bridge power" before full grid interconnection is finalized, allowing projects to commence operations sooner. The technology also supports facilities connected directly to the grid by using advanced software controls to adjust power consumption in response to grid stress, effectively turning data centers into responsive grid assets rather than passive loads. Emerald AI’s Conductor platform is central to this strategy, coordinating flexibility between compute operations and onsite resources like batteries to protect service quality while reducing infrastructure costs associated with peak demand sizing. Industry leaders emphasized the necessity of integrating energy, compute, networking, and cooling into a unified architecture for the intelligence era. Constellation CEO Joe Dominguez noted that the industry faces a supply problem driven by peak demands rather than total capacity shortages, while AES CEO Andrés Gluski affirmed that flexible AI infrastructure can function as a vital grid asset. The partners have already tested this power flexibility model across five commercial data centers globally and plan to deploy DSX Flex at commercial scale later in the year at NVIDIA’s research center in Virginia. The broader goal is to accelerate large-scale AI infrastructure deployment, support faster utility connections, and expand economic activity linked to energy investment and US energy capacity.