France needs national AI auditing platform

CNIL, the French national commission on ICT and liberty, wants France to set up a national platform for auditing algorithms. According to CNIL, AI systems need to be designed “in the interests of human freedom”.

CNIL (Commission nationale de l’informatique et des libertés) is the official French regulatory body on data privacy. The organisation sees “substantial ethical matters associated with the use of algorithms and AI” such as biases, algorithmic profiling, autonomous machines and undermining our human identity.

The findings by CNIL are the result of a public debate in France launched in 2017. Around 3.000 people took part in 45 debates set up by 60 partners (research centres, public institutions, trade unions, think tanks and companies).

The French public debate led to two founding principles to put artificial intelligence at the service of humans. First the principle of loyalty. “An algorithm – whatever it deals with personal data or not – should be loyal towards its users, not only as consumers but also as citizens, or even as communities or as an entire society”, according to CNIL.

The second principle is that of continued attention and vigilance. “It applies to every single stakeholder (designers, businesses, end users) involved in “algorithmic chains”. This represents a direct answer to some significant challenges the digital society is facing with AI.”

Apart from the national platform for auditing algorithms CNIL recommends designers and professionals in the field of AI together with citizens should be educated in the subject of ethics and AI.