What can artificial intelligence do for fairness and equality?

Challenges

ARTIFICIAL INTELLIGENCE IS A SOURCE OF PROGRESS
Artificial intelligence will change many things in our lives. For example in Health, better diagnostics and treatments may be achieved by applying advanced machine learning techniques to high-dimension quality medical datasets.
BUT THERE IS BIAS IN DATA AND ALGORITHMS
Algorithms are trained on historical data.  However, data can contain bias against people of different race, gender, age, or social background. Algorithms will not only reproduce but amplify this learnt bias, producing potentially discriminatory outcomes for entire groups of people.
 

Raising awareness and advancing best practices on Fairness and Equity In AI

 
What is Justice in Artificial Intelligence?

In partnership with Ecole Normale Supérieure, we are launching a visiting Chair in AI and Justice,
Dr. Kate Crawford, co-founder of the AI Now Institute, professor at NYU, and researcher at Microsoft, will be the inaugural chair holder.
She gave an introductory lecture on September 18th at ENS.

Our decisions have always been biased. Why would it be different for algorithms?

We have launched a partnership with TelecomParis researchers in computer science and economics who
then produced an excellent article « Algorithms: Bias, Discrimination and Equity »

What can we do to ensure algorithms are equitable and unbiased?

We are delighted to work with Institut Montaigne and have launched a working group and a series of auditions of experts on bias in algorithms.  We will be producing concrete recommendations for action, for political decision makers  and business leaders.

 

Our first projects are in neuroscience

MS BIO PROGRESS
Bio-statistics to fight Multiple Sclerosis

Dr Violetta Zujovic, Sclerose en Plaques / Multiple Sclerosis, ICM

BRAIN AT SCALE
Machine Learning to cure neurogenerative diseases, such as Alzheimer's

Olivier Colliot, Brain@Scale, ICM & CNRS

"There are moments when one wishes to give and share  …but for which cause and how can one have an impact ?  Excited by the possibilities offered by Data, I became concerned by its potential biases. With Tanya, Lawrence, Pascale and  help from the Fondation de France, I created Fondation Abeona."
 

How everything started