Unlocking the potential of AI and Generative AI in European Smart Grids – a strategic position paper and guide for action is the title of the new European Commission policy paper on Artificial Intelligence (AI) and energy systems, with INESC TEC as a coordinating entity – represented by researcher Ricardo Bessa. The document is available on the European Union’s website.
Designed under the European Technology and Innovation Platform for Smart Networks for the Energy Transition (ETIP SNET) – established within the framework of the European Union’s new integrated roadmap of the Strategic Energy Technology Plan (SET Plan) – the paper states that AI will be decisive in ensuring more efficient, resilient and transition-ready power grids, ensuring a responsible, secure and well-coordinated adoption.
The report concludes that AI is already transforming the way power grids are planned and operated, enabling more accurate forecasting of generation and consumption, congestion and fault management, predictive maintenance of infrastructures, faster response to disturbances, and greater efficiency with reduced operational costs.
“In this document, we present AI as a critical component for the integration of renewables, electric vehicles and active consumers in an increasingly decentralised system,” explained Ricardo Bessa, who leads research in the field of Power & Energy Systems at INESC TEC.
The paper also recognises the potential of generative AI for creating synthetic data when real data are scarce, supporting complex decision-making by operators, and improving human–machine interaction in control rooms.
“However, one of the key points highlighted regarding generative AI concerns the warning about the use of generic language models, which are not suitable for critical infrastructures without significant adaptation. What the document advocates is the development of AI (and language) models specifically designed for power grids, trained with data from the European energy sector,” clarified Ricardo Bessa.
The application of AI to critical infrastructures raises several challenges, ranging from cyberattacks and digital vulnerabilities to opaque or hard-to-explain decisions, privacy and data usage risks, and a shortage of specialised skills.
Hence, the report emphasises that trust in this area can only be ensured through human oversight, algorithmic transparency, robust data governance and compliance with the European AI Act.
“One of the most significant words of warning in this document is that the expansion of AI itself – particularly about data centres and large-scale models – is becoming a systemic challenge for the energy sector. For this reason, the policy paper argues that digitalisation and European policies must be planned together, that there is a strong need for energy-efficient AI, and that data centres should be wisely integrated into grids,” explained the researcher.
In addition to setting out a strategic position on AI and energy systems, the document is also intended as a guide for action. To this end, it presents a strategic roadmap with three concrete actions: short-, medium- and long-term.
In the short term – within a two-year implementation horizon – the paper calls for regulatory clarification, improved access to data, pilot-projects and skills development. In the medium term (two to five years), the roadmap foresees scaling up tested solutions, harmonising standards across countries and strengthening digital infrastructures. Finally, in the long term, it advocates the full integration of AI into power grids and European leadership in responsible AI for the energy domain.
In March 2026, the European Commission will adopt a strategic roadmap for digitalisation and AI in the energy sector, defining how to harness the potential of digital and AI technologies for the energy system while simultaneously mitigating their risks.
News, current topics, curiosities and so much more about INESC TEC and its community!
