Episode 23: The Hybrid Cloud Forecast Series – Outlook: AI Ethics
In this episode of the Hybrid Cloud Forecast series, Andre talks with Francesca Rossi, IBM Fellow and AI Ethics Global Leader from IBM Research.
Francesca shares the main goal of AI Ethics is to ensure that Artificial Intelligence advances in a way that is aligned to human values, in terms of fairness, transparency , and explainability, rather than creating an alternative form of intelligence. She talks about how machine learning models can inject bias into AI recommendations due to how the data is collected, how the training data is labeled, and how we make decisions during development. To counteract this bias, she describes the use of tools to detect bias in the training data, the AI model, or the output of the AI model. But just having the tooling is not enough. Diverse teams and robust training and educational material are required for developers to understand how to use and integrate the tools with their existing development processes. She reminds us that part of AI Ethics is ensuring that our clients will use the AI technology in line with our ethics and principals. This is done by employing a risk assessment process that considers the properties of the technology itself, the client’s planned use of the technology, and their history of using technology in an appropriate and responsible way. When the topic of data sharing comes up, Francesca talks about how the GDPR regulations in Europe were the very first regulations to govern how data is collected and shared. Francesca concludes by talking about ethics around the emerging field of Neurotechnology, reading and even writing data to our central nervous system and the ethical issues with this technology.
Art by Sam Smith.
Intro by Nathan Brophy.
Music “Happiness” by Mixaund https://mixaund.bandcamp.com/