The European Commission is opening to
civil society, academia and businesses in drafting a code of
conduct for providers of general purpose AI systems (GPAI).
The EU executive has decided to kick start public consultations
to gather the opinions of more stakeholders that will be part of
the upcoming process to draft a code of conduct for GPAI models,
one of the pillars of the AI Act, the new European law on
artificial intelligence that came into effect on August 1.
A group of European lawmakers, including the MEP who drafted the
AI Act, Brando Benifei, recently wrote to the Commission to
request a more inclusive process to draft the code of conduct
that is set to influence global AI governance.
The group warned the Commission that it appeared to be inclined
to initially involve only providers of the affected GPAI models,
with the risk of "actually allowing them to define concrete
practices on their own".
However, in a turning point, the European AI Office, set up
within the Commission, has now requested the contributions of a
vast array of interested parties, including the academic world,
independent experts, representatives of the sector, like
providers of GPAI models or providers integrating such models
within their AI systems, civil society organizations and public
authorities.
The consultation, open until September 10, will focus on
critical sectors like transparency, copyright rules, the
identification and evaluation of risk, the mitigation of risk
and internal management of risk.
The AI Act provides for an approach for GPAI on two levels, with
models presenting systematic risks subjected to more stringent
obligations.
ALL RIGHTS RESERVED © Copyright ANSA
