"Integrating ethics into algorithms raises titanic challenges" (Le Monde)

Two researchers from Télécom Paris, David Bounie and Winston Maxwell, describe in an op-ed piece to Le Monde the concrete solutions to bring to the risks of discrimination that platform algorithms can generate.

A growing number of examples in justice, health, education and finance show that artificial intelligence (AI) tools cannot be deployed unchecked in systems for security or access to essential resources at the risk of generalising biases that are potentially discriminatory, difficult to interpret and for which no explanation is provided to users.

The conclusion is becoming increasingly clear: AI must incorporate ethics by design. The ethical performance of the algorithm (absence of discrimination, respect for individuals…) must be included among the performance criteria, as well as the accuracy of predictions.

But embedding ethics in algorithms raises titanic challenges, for five reasons: First, ethical and legal norms are often unclear, and do not lend themselves to mathematical formulation[…] Second, ethics is not universal[…] Third, ethics is political[…] Fourth, ethics is economic[…] Fifth, ethics is temporal.

(c) author illustration Acidj via Freepik