We are seeing a widespread use of AI, in every aspect of our lives. We talk to virtual assistants, we unlock phones with our faces, we receive suggestion on movies or dresses from systems that know much better than us what we like. As every complex and powerful technology, AI benefits come with a cost. The use of intelligent machines, indeed, not only introduces new forms of issues, but also widen our attack surface, and is triggering huge philosophical debates. To use AI with care, thus, we should re-think the way we test and monitor our applications for fairness and robustness, the way we secure them against adversary attacks, we should find ways to explain their reasoning, and, as our major challenge, we are all called to define an ethics that is universal, systematic and 'codable'.
Speaker: Najla Said, Data Science Squad Manager, IBM