On Thursday 2 November, at the Bletchley Park AI Safety Summit, the President of the European Commission, Ursula von der Leyen, stressed the importance of a culture of responsibility at the heart of an effective system of governance for artificial intelligence (AI).
“We need to establish AI safety standards that are accepted worldwide”, she said.
Ms von der Leyen called for a framework for understanding and mitigating the risks associated with highly complex AI systems based on four pillars.
In her view, we need a thriving, independent scientific community with the means to evaluate AI systems. It is also necessary to develop internationally recognised procedures and standards for testing the safety of AI. In addition, every significant incident caused by errors or misuse of AI must be reported and followed up. Lastly, we need an international alert system supplied by trusted flaggers.
“These four pillars should constitute an effective system of governance. At the heart of it all, there needs to be a culture of responsibility” on the part of private players and public authorities, stressed the President of the European Commission.
European legislation on artificial intelligence (‘AI Act’) is currently being negotiated between the European institutions. This legislation includes, in particular, the creation of a European AI Office, which, according to Ms von der Leyen, should have a global vocation and be open to cooperation with similar entities around the world. (Original version in French by Camille-Cerise Gessant)