In her recent 12 Bytes, Jeanette Winterson is refreshingly optimistic, supposing that any future AI will be at least “unmotivated by the greed and land-grab, the status-seeking and the violence that characterises Homo sapiens”. As the computer scientist Drew McDermott suggested in a paper as long...

Could AI turn on us, or is natural stupidity a greater threat to humanity?

Ever since Garry Kasparov lost his second chess match against IBM’s Deep Blue in 1997, the writing has been on the wall for humanity. Or so some like to think. Advances in artificial intelli...

Some confusion is caused by two very different uses of the phrase artificial intelligence. The first sense is, essentially, a marketing one: anything computer software does that seems clever or usefully responsive – like Siri – is said to use “AI”. The second sense, from which the first borrows i...

Machine learning works by training the machine on vast quantities of data – pictures for image-recognition systems, or terabytes of prose taken from the internet for bots that generate semi-plausible essays, such as GPT2. But datasets are not simply neutral repositories of information; they often...

How do we get there from here, if we want to? Modern AI employs machine learning/deep learning: rather than programming rules into the machine directly we allow it to learn by itself. In this way, AlphaZero, the chess-playing entity created by the British firm Deepmind (part of Google), played mi...

In his Life 3.0: Being Human in the Age of Artificial Intelligence, meanwhile, the physicist Max Tegmark, co-founder of the Future of Life Institute (it’s cool to have a future-of-something institute these days), emphasises the problem of “value alignment” – how to ensure the machine’s values lin...

His 2014 book Superintelligence is seminal. A real AI, it suggests, might secretly manufacture nerve gas or nanobots to destroy its inferior, meat-based makers. Or it might just keep us in a planetary zoo while it gets on with whatever its real business is.AI wouldn’t have to be actively maliciou...

So-called “AI” is already being used in several US states to predict whether candidates for parole will reoffend, with critics claiming that the data the algorithms are trained on reflects historical bias in policing. Computerised systems (as in aircraft autopilots) can be a boon to humans, so th...

Will robots take all the jobs? That very framing passes the buck because the real question is whether managers will fire all the humans. The existential problem, meanwhile, is this: if computers do eventually acquire some kind of god‑level self-aware intelligence – something that is explicitly in...

So it turns all the matter on Earth into paperclips, having first disabled its off switch because allowing itself to be turned off would stop it pursuing its noble goal of making paperclips. 

That’s an example of the general “problem of control”, subject of AI pioneer Stuart Russell’s excel...

Other observers, though, remain phlegmatic. In Novacene, the maverick scientist and Gaia theorist James Lovelock argues that humans should simply be joyful if we can usher in intelligent machines as the logical next stage of evolution, and then bow out gracefully once we have rendered ourselves o...