In past blogs I’ve looked at the development of robots and machines, even speculating on what might constitute being human, if we keep replacing bits of ourselves with machines. One thing is clear to me, ever cleverer machines will happen and they will take over more and more of what have previously been human occupations. Because they can do it better, more accurately, more efficiently, quicker, safer than we can. What’s more they don’t need paying, they work 24 hours a day and don’t take holidays.
It’s happening now. Amazon are looking at replacing humans in their warehouses with robots. They’re also investigating drones to deliver them. Driverless cars are moving out of science fiction into reality. Aeroplanes most of the time are not actually flown by the human pilots.
Now it doesn’t take a huge imagination to extrapolate forwards to a world where large parts of manufacturing are almost totally the realm of robots and machines (we are part way there already). Then you order on-line and delivery is by drone. All with little or no human intervention. If all this is possible how many other occupations are ripe for machine take-over?
OK, but you still need humans to design things, invent things don’t you? Well, I’m not so sure. Perhaps for now ‘blue sky’ thinking is beyond machines, but we are working on artificial intelligence. However, most inventions / innovations are not the result of one big ‘Eureka’ moment that comes out of nowhere, but the assimilation and accumulation of what has gone before to come up with something new. Even for ideas like evolution, Darwin brought it all together and made sense of it, but the ground work had already been laid by other scientists. You could set ‘intelligent’ (I’m not defining what that means) machines with the task of accumulating vast amounts of information in related fields of work and extrapolating potential new outcomes. Who knows what they might come up with.
I know this may sound a bit random, but it’s basically how most of the pharmaceutical world operates now. You take a chemical bank of hundreds of thousands of chemicals and run them through a test, a ‘screen’ for particular properties and pick the best and work with those. So why wouldn’t this work in other areas? Surely that’s what computers do best, process vast amounts of data. The only difference I see is that they would then have to recognise the possible positive outcomes. But to me that doesn’t seem such a huge leap.
What I’ve tried to do above is show that even in areas where it seems that superficially machines, computers can’t compete with humans, maybe they can. I’m not saying that some sort of human intervention isn’t needed, it’s just how much? I know initially people will have to set these programs running, but after that? Also I’ll bet that if I can imagine this there are people working on these very things (if you know, let me know.)
When I started writing this blog I thought it was a fairly straight forward question to ask. Now I realise it is one huge can of worms and I ‘m going to need at least one more blog. The obvious question is if machines are taking over more and more of what humans currently do, what will we be doing? So next time I’ll look further at what humans might be doing / not doing in the (perhaps none too distant) future.
Opinions always welcome
Ian Martyn