‘A software engineer can create an army of AI-powered bots, each pretending to be a different person, promoting content on behalf of political or commercial interests. Unlike broadcast propaganda or direct marketing, this approach also uses the self-reinforcing qualities of the algorithm to learn what works best to persuade and nudge each individual.’
Three AI thinkers say smart algorithms should be treated like drugs and scrutinized for harmful effects before being released to the public. Writing in WIRED, Businessman Olaf Groth, think-tank leader Mark Mitzberg, and prominent AI professor Stuart Russell call for consumer protection measures similar to those of the FDA, which controls the release of prescription drugs in the USA.
They say that at scale, the capabilities of learning-based algorithms create ‘an unprecedented force multiplier’ and can be used to promote the interests of an interest group or individual.
Olaf Groth is a professor at Hult Business School and an entrepreneur. Mark Nitzberg is executive director of the Center for Human-Compatible AI at UC Berkeley. Stuart Russell is a computer science professor at UC Berkeley. He is author (with Peter Norvig) of the landmark textbook about artificial intelligence and of Human Compatible: AI and the Problem of Control (2019). Groth and Nitzberg are coauthors of Solomon’s Code: Humanity in a World of Thinking Machines (2018).
Groth, Nitzberg, Russell: AI Algorithms Need FDA-style Drug Trials
WIRED | August 15, 2019