Daisy Cross
Head of Future Retail Markets, Energy UK
Artificial intelligence in the energy sector isn’t about fully autonomous power stations yet, but it’s already optimising products, services and operations across the industry.
As the energy industry innovates, the system continues to become smarter, helping offset some of the costs and risks faced by industry and its customers.
AI in energy is already being used
To begin with, AI is already giving us a better idea of what real-time supply and demand looks like across the network, with more accurate and efficient forecasting, getting electricity to the right place at the right time. These benefits will only multiply as more renewable electricity gets added into the grid by solar and wind farms (NESO predicts AI could improve solar forecasting accuracy by 33%).
Some AI energy uses are more abstract. For example, National Grid has used AI to analyse whale migration patterns alongside six years of vessel data, processing 4.7 billion data points and 775 million kilometres of vessel traffic to establish whether wind farm operations were disturbing whales’ natural behaviour (they were not).
Closer to home, suppliers are competing to use generative AI to improve customer service experiences, such as triaging contacts and prompting follow-ups. Anonymised real-time meter data is being optimised by AI to proactively identify customers who might be vulnerable, so they can get support as quickly as possible.
Suppliers are competing to
use generative AI to improve
customer service experiences.
Pro innovation approach
These benefits are informing the Government’s wider vision of the UK as a globally competitive AI-friendly powerhouse. The UK AI market is already worth more than £70 billion, making it the third-largest AI market in the world.
However, there is scope for growth. This is why the energy sector supports the Government’s direction to Ofgem and other regulators to adopt a pro-innovation approach to AI rulemaking, so customers and the system are protected and our competitiveness is maintained, but not in a way that slows progress or investment. So, useful measures such as the incoming AI technical sandbox — a controlled environment in which energy providers can safely test and evaluate AI models and behaviours — should be welcomed.
