Unpredictability

The recent revival of artificial intelligence, combining machine learning and big data, promises to reveal in advance what will happen in the future. Algorithmic forecast, however, is very different from the idea of prediction that has established itself in modern society since the eighteenth century, oriented and guided by the idea of probability and its calculus. Algorithmic predictions, that are individual, contextual, and performative, can lead to self-fulfilling circularities but also to pre-emptive policies, which limit future possibilities. Their problem is not so much the missing data, but rather the available data and the way they affect our attitude towards the future. Even when correct, algorithmic prediction can prove ineffective and the future remains unpredictable.