AI, or not AI? by Neil McGowan
The last few weeks have been a hectic mess. I’ve not been writing as much as I’ve been too busy dealing with the vagaries of real life, but did manage to revisit an idea I’d sketched out some years ago. The seeds of this story hark back around seven or eight years, and were planted when I heard Professor Stephen Hawking talking about the dangers of artificial intelligence. (Having seen the output of most of the machine learning models so far, I think there’s quite a way to go yet; I usually refer to it as ‘artificial idiocy’ – yes, I’m a curmudgeonly old technology sceptic – as most of the time it seems too easy to find faults either in the system itself, or the training data used ; you don’t have to look far to find stories illustrating how biased facial recognition systems can be, for example.) Over the next few days, there was a raft of stories across various media, either proclaiming AI as the salvation of humanity, or the beginning of the end. But one thing struck ...