Academics to chair drafting the Code of Practice for general-purpose AI

A range of academics, from Turing award-winner Yoshua Bengio to PhD candidates, have been named chairs and vice-chairs of working groups that will draft a Code of Practice on general-purpose artificial intelligence (GPAI), according to a Monday (30 September) Commission press release.

For providers of general-purpose AI systems like ChatGPT, the AI Act relies heavily on the Code of Practice, which will detail what the Act's risk management and transparency requirements would entail in practice until standards are finalised, sometime in 2026.

"Chairs and vice-chairs play pivotal roles in shaping the first general-purpose AI Code of Practice," the Commission said in a press release.

Last week, three influential MEPs questioned whether the working group leaders would have international expertise and the timing of the list's release.

The Commission published the list just hours before the drafting process of the Code formally started with an online kick-off plenary at 3 pm today (30 September) according to a Commission email seen by Euractiv.

The Code's first draft will be presented around 3 November, and the initial GPAI provider workshop is scheduled for mid-October, the Commission said during the plenary, according to a source.

Kai Zenner, digital policy advisor to MEP Axel Voss (Germany, EPP), one of the three who sent questions to the Commission about the chairs and vice-chairs, said he was happy to see international expertise in the list but that it was a "pity" the announcement came after the Parliament's intervention, in a LinkedIn post.

The list is international and highly skewed towards academia and research institutions, with four professors, two PhDs, three PhD candidates and three researchers among the 13 appointed chairs and vice chairs.

The names

One name that attracted positive attention was professor Yoshua Bengio, who will chair the group for technical risk mitigation. He is often referred to as one of the three 'godfathers of AI' because his work on deep learning in the 1980s was fundamental to AI as we know it today.

“Having Yoshua Bengio using his oversubscribed time to chair a Code of Practice working group demonstrates the importance that the EU AI Act represents globally as the leading regulation for AI risk management,” AI risk management expert and CEO of SaferAI Simeon Campos told Euractiv.

Bengio will be supported by two vice-chairs: Daniel Privitera, lead writer of the International Scientific Report on the Safety of Advanced AI, and Nitarshan Rajkumar, PhD candidate and previous senior policy advisor for the UK government.

Director of the European Laboratory for Learning and Intelligent Systems (ELLIS) Alicante Nuria Oliver and law professor Alexander Peukert will co-chair the group working on transparency and copyright-related rules.

The group's vice-chairs are computer science PhD candidate Rishi Bommasani and law professor Céline Castets-Renard.

Matthias Samwald, associate professor at the Institute of Artificial Intelligence at the Medical University of Vienna, will chair the working group on risk identification and assessment, with support from vice-chairs Marta Ziosi, a postdoctoral researcher, and Alexander Zacherl, a former Systems Designer at Google DeepMind.

Finally, Marietje Schaake, a former MEP and fellow at Stanford's Cyber Policy Centre, will chair the working group on internal risk management and governance of general-purpose AI providers.

She will be helped by vice-chairs Anka Reuel, PhD candidate at Stanford, and Markus Anderljung, a researcher at GovAI, who has participated in several international AI governance initiatives.

[Edited by Eliza Gkritsi/Martina Monti]