The Hidden Dangers of Black Box AI 🚨 | Javier Viaña at TEDxBoston

"The excessive use of “black box” Artificial Intelligence is threatening our ability to supervise machines.

If we do not make soon a radical change in the way Neural Networks are designed and function, we are going to face a global storm of uncertainty and lack of control over the deployed AI.

The field of eXplainable Artificial Intelligence (XAI) arose to mitigate this lack of explainability, but there is still much that needs to be done if we want to fully understand the inner logic of AI.


AI, Algorithm, Computers, Machine Learning, Technology Javier Viaña is currently a Postdoc Associate at the MIT Kavli Institute of Astrophysics and Space Research. He obtained his Ph.D. in eXplainable Artificial Intelligence applied to the Aerospace Engineering at the University of Cincinnati and spent his last doctoral year at MIT. His Ph.D. research revolved around the conception of novel transparent algorithms that not only provide accurate predictions but also human-understandable justifications of the results. His current research is focused on the design of deep neural network architectures that automatically detect transiting exoplanets from Kepler satellite data. Previously, Javier developed tailored AI solutions for different aerospace organizations such as Aurora Flight Sciences, Boeing, Satlantis Microsatellites, NASA, ESA, Genexia, and the Northern Kentucky International Airport. He graduated from his Bachelor's in Mechanical Engineering at the University of the Basque Country, and Master’s in Aerospace Engineering at the University of Cincinnati. His main research topics and interest include eXplainable AI (XAI), deep fuzzy networks, genetic fuzzy systems, bio-inspired evolutionary optimization, white dwarfs, exoplanets, and black holes.

He has published four books, more than 20 peer-reviewed papers, several registries of intellectual property and four patents. He has received awards such as the Best National Research Article of 2016, the “La Caixa” fellowship award, the John Leland Atwood Award of the American Institute of Aeronautics and Astronautics, a letter of recognition from HH. MM. King Felipe VI and the Royal House of Spain, the Professor R. T. Davis Memorial Award for Academic Performance in Computational Sciences, the Airport Cooperative Research Program Award from the Federal Aviation Administration and The National Academy of Sciences, and 3 Outstanding and Best Paper Awards from the North American Fuzzy Information Processing Society’s (NAFIPS) Annual Conference. This talk was given at a TEDx event using the TED conference format but independently organized by a local community. Learn more at https://www.ted.com/tedx