Emerald AI trials in London show AI factories can ease UK grid strain
Executive summary
Emerald AI, an NVIDIA partner, has demonstrated how its 'Conductor' platform can enable AI factories to become flexible power grid assets. Trials, including one recently held near London, showed AI infrastructure can reduce power demand significantly and rapidly, thereby assisting national grid stability during peak loads or unexpected events. This technology aims to unlock untapped grid capacity for AI deployment.
Reporting based on nvidia.com
Why it matters
For the UK, where concerns about energy supply and grid stability intersect with rapidly growing AI infrastructure, Emerald AI's technology presents a potential solution to mitigate the energy demands of AI factories. By allowing AI data centres to dynamically adjust their power consumption, it could help integrate more AI capacity without necessitating immediate, massive upgrades to the existing grid infrastructure, while also supporting the National Grid in managing fluctuations.
Sector impact
Analysis by AI Energy Intelligence UK
The technology could significantly alter the profile of electricity demand from AI data centres. Instead of being a constant, inflexible load, AI factories could become responsive assets, reducing demand by up to 40% in under a minute. This flexibility could help manage peak demand periods and other grid stresses, potentially alleviating the need for backup generators during energy usage spikes.
By transforming AI factories into 'power grid assets' that can flexibly respond to grid demands, the technology contributes to enhanced energy security. It enables rapid 'emergency curtailment' to maintain grid resilience, as demonstrated by shedding approximately 30% of a data centre’s load within 30 seconds. This capability could prevent blackouts and stabilise the grid during unforeseen events or prolonged stress.
For businesses operating AI data centres, adoption of Emerald AI's platform could lead to cost savings through more affordable power bills, as the system utilises existing grid capacity more effectively. It also allows AI factories to be connected faster without triggering demand surges, potentially accelerating the deployment of new AI infrastructure. The collaboration between NVIDIA, Emerald AI, and partners like Nebius highlights a growing focus on sustainable and grid-responsive AI operations.
While not directly affecting individual consumers, the technology has the potential to make power bills more affordable for energy consumers by utilising existing grid capacity more efficiently. A more stable and resilient power grid, less prone to energy usage spikes and backup generator activation, indirectly benefits all consumers by contributing to reliable and potentially lower-cost energy supply.
Key statistics
Figures as reported by nvidia.com. See original source for context.
Quotations
"The power grid is like a large-scale freeway that only faces rush hour two times a month. That’s how underutilized our power grid is."
"Together with NVIDIA, we're proving that AI compute can be power-flexible, a paradigm shift with a massive prize: unlocking 100 GW of untapped power, grid capacity, and resolving AI's energy bottleneck, while promoting reliable and clean power grids."
Long-term implications
The long-term implications suggest a fundamental shift in how AI infrastructure interacts with national power grids. If widely adopted, power-flexible AI factories could integrate more seamlessly into existing energy systems, potentially unlocking significant untapped grid capacity (e.g., 100 GW in the US). This could enable more rapid and sustainable growth of AI technologies globally, while promoting grid reliability and potentially influencing future energy policy and infrastructure investment.
Frequently asked questions
What is the Emerald Conductor platform?
The Emerald Conductor platform is a solution developed by Emerald AI that transforms AI factories into power grid assets. It allows them to flexibly use power, utilising untapped capacity and reducing demand during peak grid loads.
How effective is the Emerald Conductor platform at reducing power demand?
In trials, the Conductor platform was able to reduce power demand by up to 40% in under a minute while maintaining critical workloads, and shed approximately 30% of a data centre’s load within 30 seconds for emergency curtailment.
Has this technology been demonstrated in the UK?
Yes, a five-day demonstration took place at Nebius' new AI factory outside of London, United Kingdom. This trial used a 96-GPU NVIDIA Blackwell Ultra AI cluster combined with the Emerald Conductor platform.
What impact could this have on the UK's National Grid?
The platform demonstrated 100% alignment with National Grid's power targets, validating that high-performance AI infrastructure can act as a precise, grid-responsive asset. It can support load-reduction requests for up to 10 hours, aiding sustained flexibility during prolonged grid stress.
Explore related tools
Original source
This story summarises reporting from nvidia.com. Read the original for full context.
Read on nvidia.comRecommended reports
Independent UK research that expands on the themes in this story.
The definitive UK view of AI's electricity, grid and infrastructure impact in 2026. An independent, evidence-based report on demand growth, data centre build-out, AI Growth Zones and the trajectory to 2035 — for UK policymakers, operators and investors.
Read the reportThe environmental cost of AI-generated answers versus conventional search, benchmarked for UK decision-makers. An independent, evidence-based comparison of electricity, water and carbon per query across Google, Gemini, ChatGPT, Copilot, Perplexity and Claude — with forecasts to 2035.
Read the reportBottom-up model of AI-driven electricity growth across UK regions to 2030, with sensitivity to model efficiency and adoption.
Read the reportGet the UK AI Energy briefing
Analysis on AI, electricity and the UK grid, straight to your inbox.