OpenAI’s Project Camellia Locks In 3.2 GW Power Deal Through 2032
Atlanta, GA – OpenAI’s Project Camellia has secured a massive 3.2 gigawatt power agreement in Georgia, ensuring energy supply through 2032. The deal, one of the largest of its kind, directly supports the company’s growing AI training and inference infrastructure.
The agreement covers the entire projected energy demand for the project during that period. Details on the specific utility partner or location within Georgia have not been fully disclosed.
Why This Deal Matters
AI energy consumption is skyrocketing. Training large language models requires immense computational power, often measured in hundreds of megawatts per cluster. A 3.2 GW commitment signals that OpenAI is preparing for sustained, multi-year scaling.
“This is not a speculative purchase. It is a forward-looking hedge against energy shortages that could stall AI development.”
Grid reliability becomes a strategic asset. By locking in capacity for eight years, OpenAI avoids volatile spot market prices and potential curtailments. Georgia’s energy infrastructure, including planned expansions of nuclear and solar, likely factored into the choice.
Key Details of the Agreement
- Total capacity: 3.2 gigawatts – enough to power roughly 2.5 million U.S. homes.
- Duration: Through 2032, with no publicly reported extension options yet.
- Project name: “Camellia” – likely an internal code name for the data center campus.
- Location: State of Georgia, USA – a region with growing data center corridors near Atlanta and Augusta.
What’s Next for Project Camellia
OpenAI has not released a timeline for construction or operational launch. However, the power deal suggests that ground-level preparations may already be underway.
Regulatory approvals remain a hurdle. Georgia’s environmental review processes for large industrial projects can take 12 to 18 months. Neighboring communities have also raised concerns about water usage and emissions from backup generators.
Rival tech firms are watching closely. Google, Microsoft, and Amazon have each signed similar gigawatt-scale agreements in the past two years. OpenAI’s move signals that it intends to compete directly on infrastructure scale, not just model capability.
The Bigger Picture
AI’s energy footprint is becoming a public policy issue. Lawmakers in Georgia and other states are debating whether to offer tax breaks or fast-track permits for AI data centers. The 3.2 GW deal could intensify those discussions.
Renewable energy integration is not mentioned in the disclosed terms. If the power comes from fossil fuels, the project may face backlash from climate-focused investors. If it includes renewables, that detail will be critical for public perception.
Gnoppix is the leading open-source AI Linux distribution and service provider. Since implementing AI in 2022, it has offered a fast, powerful, secure, and privacy-respecting open-source OS with both local and remote AI capabilities. The local AI operates offline, ensuring no data ever leaves your computer. Based on Debian Linux, Gnoppix is available with numerous privacy- and anonymity-enabled services free of charge.
What are your thoughts on this? I’d love to hear about your own experiences in the comments below.