Information and the Regulatory Landscape: A Growing Need to Reconsider Existing Legal Frameworks

Abstract

Advanced artificial intelligence (AI) systems are already being used to enhance our lives and to transform the way businesses operate. Businesses across a broad spectrum of industries are exploring the potential gains offered by AI systems. In fact, the use of AI systems is already widespread in areas such as transport, finance, defense, social security, education, policing, public safety, and healthcare. The recent explosion of machine learning technology is arguably a product of two things: “tremendous increases in computational power and enormous volumes of accumulated data.” Unsurprisingly, legal frameworks and industry-based governance regimes have failed to keep up with the newest AI. The existing gaps have led to industry attempting to fill the void, but these attempts are in their infancy and often fail to fully consider the various stakeholders impacted by the ubiquitous gathering and corresponding use of data.

Consider the recent news splash concerning Google’s advertising program, which is once again under fire for its use of highly secretive gathering, storing, and using of highly sensitive data. According to the Electronic Privacy Information Center (EPIC) Google is gaining access to “highly sensitive information — the credit and debit card purchase records of the majority of U.S. consumers — without revealing how they got the information or giving consumers meaningful ways to opt out.” And, as is often the argument against the use of black box algorithms, EPIC asserts the “search giant is relying on a secretive technical method to protect the data.” Of course, this is not the first — nor the last — criticism of big businesses’ use of black box algorithms.

This Article seeks to further debates previously asserted by the author by examining the issue in light of the recent surge in attention algorithms are drawing, primarily because their use has grown exponentially.