Artificial intelligence (AI) is increasingly being used in sectors to support the delivery of wide societal benefits, from medical advances to mitigating climate change, and to enable businesses to develop new, advanced solutions. Indeed, the UK AI market is already worth more than £70 billion, making it the third largest AI market in the world.
In the energy sector, AI can be used to develop innovative products and services that support better outcomes for households, communities and businesses, and enable a smarter, digital energy system at lowest cost to them. Already, many companies are using AI tools to transform how they interact with customers, develop smart solutions, and support decarbonisation goals.
However, there is also an opportunity for innovation, with only 6% of energy companies rating their maturity as 5 out of 5, in a survey by Mesh-AI in 2023. This compares to 3% for financial service organisations, and perhaps unsurprisingly, with technologies companies far in the lead at 18%.
The UK Government sees AI as a unique chance to “turbocharge” its Plan for Change and aims to build on the UK’s position as a global leader. Earlier this year, it endorsed the AI Opportunities Action Plan, which has 50 recommendations to harness the prospects of AI, including improving data capabilities and access, reforming regulation, and driving adoption across both public and private sectors.
The Department for Science, Innovation and Technology has also produced a white paper on AI regulation, setting out expectations for UK regulators to adopt a pro-innovation approach. In response to this, Ofgem has published its strategic approach to AI, and developed good practice guidance for the sector this year. Ofgem is also running regulatory laboratory sessions on hypothetical or real AI use cases, with the next session in October, and is currently seeking views on creating an AI technical sandbox, which will provide a secure, controlled environment for energy sector participants to safely design, test, and evaluate AI models, datasets, and system behaviours.
In light of this developing landscape, Energy UK has been further exploring the opportunity AI use presents for the energy industry and its customers, and how to optimise the routes which provide value to people and businesses. This has included convening a cross-membership AI working group on upcoming policy and holding events where stakeholders can share insights.
While we don’t have all the answers, it is likely that AI will only grow in use and interest for the energy industry in the coming years. As an introduction, here are a few use cases which highlight the unique opportunities of AI adoption.
Targeting consumer support
One area where AI is already being used is to support customers in need, for example the uZero platform. This platform identifies areas living in fuel poverty, by using anonymised real-time smart meter data from the Data Communications Company’s network alongside other cross-sector datasets.
This allows Ofgem, government, housing associations, energy providers and social care providers to better understand where there are high proportions of households experiencing vulnerability, so they can tailor solutions appropriately. For example, working with The Wise Group, Octopus Energy has used this approach to more easily identify households across its service areas that are eligible for the Warm Home Discount.
Improving customer service
Organisations are using AI to enable more efficient and targeted customer service, to increase customer satisfaction and engagement. This includes using AI tools to support staff to triage requests, provide prompts to ensure all consumer questions are addressed, and detect customers who are at risk. For example, in Energy UK’s Vulnerability Commitment Good Practice Guide 2024, ScottishPower and EDF highlighted how they are using AI to enable a deeper understanding of the underlying reasons for customer contact and to help identify customers experiencing vulnerability.
Citizens Advice and the Incubator for Artificial Intelligence have also been trialling a customer service AI copilot service, to enable its advisors to quickly locate and share information from reliable sources. In addition to generating high quality and detailed messages faster, advisors using this tool were more than twice as likely to feel confident giving advice, and more than one-and-half times as likely to feel able to resolve the issues being raised by clients.
Optimising the grid
AI technologies are helping to optimise the energy grid through more accurate energy supply and demand predictions. With an increasing proportion of renewable sources on the grid, accurate forecasting of solar and wind generation is essential to operating the system economically and efficiently. To maximise efficiencies, NESO has found that combining several AI models could produce a solar forecasting system 33% more accurate than traditional methods.
The energy industry also recognises that there is an increasing demand for data centres (often used for complex AI applications), and for adequate investment in new low-carbon power sources to meet this new need. Energy UK has recently released a report exploring this, setting out clear recommendations on how the UK should streamlines processes to deliver the potential for investment and growth across low-carbon technologies, data and AI.
From more effective customer support to optimising the grid, AI is already delivering better outcomes in the energy industry. Stay tuned for further insights into the AI-powered energy landscape, as we will unpack these advances in more detail, and explore other emerging challenges and opportunities in upcoming blogs and events.
