Intelligence artificielle et métrologie
The BIPM brought together experts from various National Metrology Institutes (NMIs) to discuss topics ranging from quantum technology measurement standards to the FAIR principles for data management in metrology – during our recent two-day NMI Directors workshop.
One of the most popular sessions was on the intersection of artificial intelligence (AI) and metrology, focusing on how precise measurement can enhance the control and quality of AI systems.
In particular, Dr Agnès Delaborde from the Laboratoire National de Métrologie et d'Essais (LNE) presented insights into “Enhancing control of Generative AI through metrology,” addressing bias, uncertainty and the importance of certifying AI models. By applying metrological techniques, her presentation emphasized how accuracy and trustworthiness can be reinforced in AI outputs.
Prof. Dr Tobias Schäffter from the Physikalisch-Technische Bundesanstalt (PTB), also led a compelling discussion on “Data quality metrics and reference data for AI,” highlighting the critical need for high-quality datasets and standardized metrics to improve the reliability of AI systems.
Both presentations underscored the essential role metrology has, and will continue to, play in shaping the future of AI and ensuring its integration into key sectors with the highest standards of accuracy and integrity.
- Watch Dr Agnès Delaborde’s full presentation here.
- Watch Prof. Dr Tobias Schäffter’s full presentation here.
- Access all of the presentation materials for this meeting here.