Insight 24: The Role of AI across Battery R&D and Manufacturing
Summary
Artificial intelligence (AI) is accelerating scientific discovery across the economy, with potentially far-reaching implications for battery innovation. Its applications span materials discovery, cell and pack design, manufacturing and diagnostics. In battery development and production, AI can shorten development timelines, improve performance, and support advances in both established and next-generation technologies.
Batteries and AI are central to the UK’s Modern Industrial Strategy, and the UK already has world-leading capabilities in both fields. Realising the full benefits will require closer integration of data and AI tools throughout the battery value chain, underpinned by high-quality data, trusted models and interdisciplinary collaboration.
Focus of the Insight
Key points from this Insight:
- AI is now widespread across industry: a recent McKinsey survey found that 88% of organisations use AI in at least one business function, up from 55% in 2023. However, full integration remains rare, with only 7% of adopters reporting that AI is fully embedded in their operations.
- AI is a toolbox, not a single technology: AI encompasses a range of tools and approaches designed for specific tasks. These tools rely on distinct datasets, assumptions and objectives; a model developed for materials discovery may not be suitable for manufacturing optimisation or battery diagnostics.
- AI is being applied throughout the whole battery value chain: Applications range from screening and generating candidate materials, accelerating cell and pack design, and optimising manufacturing quality and yield to improving battery diagnostics and battery management systems, and enabling automated, self-driving laboratories.
- The UK has strong and growing capability: Examples include CuspAI’s AI Materials Foundry; Imperial College London’s spin-out Polaron and the DIGIBAT self-driving laboratory; IONETIC’s AI-based battery pack design tools; Eatron and Elysia’s AI-enhanced battery management systems; Generational’s EV battery diagnostics; and Faraday Institution projects including Nextrode and FAST alongside the proposed “Faraday Engine”.
- UK policy and infrastructure are beginning to match this ambition: The UK government recognises both the scale of the challenge and the opportunity AI presents, with significant planned investment in this field. This includes the £137 million for AI-enabled scientific advances from the AI for Science Strategy and the £500 million Sovereign AI Fund. Industry also intends to develop these capabilities in the UK, illustrated by Google DeepMind’s planned UK materials-discovery laboratory, for which batteries are a named application area.
- The most transformative opportunity is AI-led experimentation: Closed-loop experimentation systems could connect materials discovery, cell design, manufacturing and performance monitoring within a single iterative process, This would shift the human role from routine experimental execution towards strategic direction, oversight and validation.
- Realising this potential is not automatic: Wider adoption and integration of AI-based approaches will depend on addressing interconnected barriers relating to data availability and quality, interdisciplinary skills, trust in model outputs, robust validation, computing infrastructure and organisational culture.
Conclusion
AI has already demonstrated its ability to accelerate R&D and improve battery manufacturing processes. It also has clear potential to strengthen the UK’s position in advanced battery research and industry. Capturing this opportunity will require sustained investment in data infrastructure, skills, computing capacity and data standardisation. Human expertise will remain essential, increasingly focused on designing, overseeing and validating capable AI systems. Given the UK’s strengths in both AI and battery research, the coming years present an important window to translate this capability into lasting scientific and industrial advantage.