SCAI INITIATIVE ON

RESPONSIBLE AI

An initiative to raise awareness of the societal impacts of Al and to promote its responsible and safe use through research, education, and outreach.

Our goal is to ensure every student understands Al's societal impact.

Let's demystify Al and promote its responsible use across society.

Trustworthy machine learning underpins responsible and sustainable AI technologies.

Research Al trust, including data protection, bias, fairness, robustness and security

Teach students about Al benefits and limitations

Explain Al's societal impact to the public through events and content

AI models are becoming an increasingly integral part of our daily lives, yet their use carries significant (direct and indirect) societal and technical implications and risks.

To address these challenges, we are launching a new initiative focused on raising awareness of the societal impacts of AI and promoting best practices for more responsible and safe use of AI.

This initiative, a collaboration between SCAI and associated members of the PostGenAI@Paris cluster, seeks to unify ongoing research and outreach efforts under a structured program.

By pooling resources and expertise from diverse institutions, we aim to spotlight the limitations and critical questions arising from AI adoption.

Our approach aspires to position us as leaders in Responsible AI, fostering impactful research, meaningful dialogue, and actionable changes in the education of future practitioners.

Our Three Pillars

The SCAI initiative on Responsible AI is built on three pillars: Research, Education, and Outreach.

Research

At the core of our initiative is the creation of a consortium of researchers in mathematics and computer science dedicated to advancing the technical aspects of AI trust.

This includes, but is not limited to, studies in data protection, algorithmic bias, fairness, robustness, and security. Through scientific events, workshops, and collaborative sessions, this pillar aims to encourage growth in the vital field of AI safety, security, and trustworthiness.

Education

As academic institutions, we recognize our responsibility to prepare future practitioners to navigate the evolving landscape of AI.

AI will inevitably become an important tool for most professions in the future, making it essential to equip students with a clear understanding of both the benefits and limitations of AI-based solutions. This pillar seeks to enhance curricula at Sorbonne Université and partner institutions by introducing courses and seminars on AI’s societal impacts, emphasizing the importance of awareness around its technical limitations.

Outreach

In a field as influential as AI, researchers must extend their conversations beyond academic circles.

It is vital to engage the public and foster awareness of the societal stakes posed by new technologies. This outreach pillar will focus on engaging broad audiences by addressing AI’s implications across various domains such as environmental, legal, security, political, and economic. We aim to host conferences and produce accessible content, such as blog posts and articles, to demystify complex topics and promote public understanding of the societal impacts of AI.