Rodney Brooks, former director of the Computer Science and Artificial Intelligence Laboratory at MIT, makes claims around the 'Seven Deadly Sins of AI Predictions'- the title of his recent article for The Technology Review. The list compiles mistaken extrapolations with limited imagination, and other mistakes which are all working in a way which distracts from productive thinking about the future. Amidst this hysteria about the future of artificial intelligence, robots and their capabilities (even to turn our jobs into theirs), is the safety net of predictions - 'the speed of deployment'.
Is time on our side? It goes back to that old saying - the saying that we will cross that bridge when things get to that certain point.
It is noted that although new versions of software are released regularly, with the likes of Facebook and Twitter updating their features almost hourly, deploying new hardware is a greater challenge.
We see this firsthand in our own lives. A conversation I was having the other day with a tech-minded car enthusiast, underlined the idea that many of the cars we are buying today (the ones that still need drivers to operate them), will most likely still be driving up and down the motorways in twenty years time. As Brooks relays, this acts to limit how soon all our vehicles will be self-driving.
Even the house I live in - it is over 100 years old, and as Brooks points out even if I were to build a new home today, one can expect that it might be around for over 100 years to come.
Put simply, it is the capital costs that keep such physical hardware around longer, even if tech is involved.
I like to think that the world is already digitally capable, but there is a still a long way to go. Near to all the innovations in AI and robotics will take far longer to be deployed, and then employed, by us all. Does this mean we have time to prepare - prepare our jobs, the industries we are in, and start saving for the next smart home or driverless car purchase? Will it all be A-OK?
Almost all innovations in robotics and AI take far, far, longer to be really widely deployed than people in the field and outside the field imagine.