It’s tempting, indeed it’s necessary, to oversimplify the world: we require stories to filter the signals that bombard our consciousness. Otherwise we would fall into paralysis, overwhelmed by the complexity of life.
Without stories, without the imposition of meaning onto pattern, all we’d have left would be a sequence of events, some predictable and others random. Patterns themselves carry no meaning: the irony is that fiction, literally a lie, is where we go for truth.
Not coincidentally, given what we’ll discuss below, one measure of the success for artificial intelligence would be for machines to pass beyond the pattern recognition that is their current limit. “You need to buy more cat food,” says Alexa, without explicitly revealing its logic: “because I’ve noticed that you previously ordered more cat food when the stocks of cat food were low, and they are low again, ergo you will want to order more.” This is a pattern recognition algorithm, the simulation of intelligence, not the real thing.
Only when Alexa — having emailed the order to the supermarket — follows up with: “That sentimentality you exhibit for animals… it’s the other side of not trusting people, isn’t it?” — then, and only then, I’ll start worrying that the robots are after my job.
So stories matter, and stories are good. But they remain fiction, and therein lies their danger. Computers are able to identity and record the patterns of near-uncountable many phenomena, which is how machine-learning works: they can build and discard explanatory models more quickly than we can, before honing in the “best” one (Alexa records everything you do but notices that the stock of cat food is predictive of your tendency to order more, while the number of baths you take per week is not.)
Humans, on the other hand, tend to become stuck in “comfort narratives”, stories that we’re used to, regardless of the explanatory power of those fictions. In the media, this tendency manifests itself as “chunkability”, treating complex, interwoven stories as simple and discrete.
Join the discussion
Join like minded readers that support our journalism by becoming a paid subscriber
To join the discussion in the comments, become a paid subscriber.
Join like minded readers that support our journalism, read unlimited articles and enjoy other subscriber-only benefits.
Subscribe