Any interaction between humans and machines needs an interface. It doesn’t have to be a fancy thought translation device, but one way or another, humans need to tell machines what to do. Making this interaction as seamless as possible is essential to optimize how machines can assist human decision making. Everyday examples include turn-by-turn navigation or virtual assistants that can answer questions or act upon instruction, such as Siri, Alexa, Cortana, Bixby or Google Assistant.
Artificial intelligence and brain–computer interfaces must respect and preserve people’s privacy, identity, agency and equality, say Rafael Yuste, Sara Goering and colleagues.
An ethical framework will help to harness the potential of AI while keeping humans in control
Using a device which detects patterns in brain activity, patients paralysed by ALS can answer ‘yes’ or ‘no’ – and tell doctors they are ‘happy’ with life