Study: AI capable of deceiving humans, poses serious risk

A new study by researchers from the Center for AI Safety in San Francisco has revealed the risks of AI’s increasing capacity to lie.