Oh no, here we go...

Prof. Stephen Hawking has joined a roster of experts worried about what follows when humans build a device or write some software that can properly be called intelligent. Such an artificial intelligence (AI), he fears, could spell the end of humanity.

And the BBC interprets that as; 'Does rampant AI threaten humanity?'

Rampant? You mean; out of control, violent, fanatical, unbridled and pervasive? I'm pretty sure that rampant milk bottles would be something to fear so, of course, rampant AI would be terrifying, but what are we really saying here? That superior intelligence is fearful? You mean the thing we aspire to, the goal that civilisation strives for – higher intelligence and greater knowledge – is something to fear? Coming from Professor Hawking, that's a mind-bender.

We need superior intelligence. We need better systems. We can no longer rely on fallible humans to protect us. We want the best possible systems watching out for us, to quadruple-check every possible outcome twenty-five thousand times faster than we can.

You are driving down a road. A car is coming towards you at a closing speed in excess of 80 miles per hour and all that’s separating you is a line of paint on the tarmac. Now who do you want in control? An old guy with a dickie ticker, a mother screaming at her kids creating mayhem on the back seat, a stressed-out businessman trying to text his quotation for a deal that will salvage his career, a teenager about to sneeze or an AI that will evaluate every possible parameter from the engine performance and the atmospheric conditions to the tyre/road cohesion coefficient and all potential peripheral incursions into the projected vector of travel? Do you want a tired engineer phoning his bookie or a computer making sure your train doesn’t pass a red light? Do you want a pilot who pulls back on the stick all the way to the cold, hard Atlantic or a system that can override our confusion and keep us alive?

But wait! They’re all going to go rampant! You can't have a super intelligent computer make decisions for you, you have to leave everything to attention-span-challenged, meat-brained humans to screw it up instead.

We need AI for more reasons than we are able to figure out. Of course we need to safeguard against harm, but we’ve been doing that since we rubbed sticks together to make fire. The problem isn’t greater machine intelligence, it’s the lack of enough intelligence to implement our innovations securely. That’s what Prof. Hawking is warning us about. That baby in the bathwater will grow up to save humanity and move us into a post-scarcity civilisation beyond our imagining – it just has to be taught how to behave.