How can we develop AI that helps, rather than harms, people?

In every technological revolution, we face a choice: build for freedom or watch as others build for control. With AI the stakes couldn’t be higher. It already mediates 20 per cent of our waking hours through smartphones, automated systems, and digital interfaces. Soon it will touch nearly every aspect of human existence. While AI promises to liberate us for higher pursuits by “extending the number of important operations which we can perform without thinking,” history – from the iron cage of Soviet bureaucracy to modern Chinese surveillance – serves as a stark warning that automation can just as easily erode our freedoms and condition us to passively accept social control.

Today’s debate about AI’s future is dominated by competing visions of control. Doomsayers, like some of those at this week’s AI Action Summit in France, advocate for strict controls (even “pauses” on all development) that would forfeit progress while inviting tyranny.