As AI has become the latest fashionable technological craze (crypto is so last year!), I have been inundated with ‘information’ about it, a lot of it is hysterical and ill-informed. The press is generally having a field day. One side is trying to frighten readers with an existential crisis about how AI will take over the world and destroy the human race, or use us as slaves. The other angle is how AI will create an utopia where it has cured all illness and created prosperity for all. There is very little realistic appraisal of the real effects it is already having and how in the medium term there will probably bit a bit of both – the balance being decided by who gets to control it.
Except for the Mastodon Fediverse I’m not a consumer of antisocial networks, but I expect that generally they are the same – except even more extreme. Surprisingly on my corner of the Fediverse (Fosstodon), which is mainly about computer/technical subjects, there has been very little discussion of the topic.
Having studied computer science at university the topic interests me, so I have watched, listened to and read quite a lot of information about AI recently and one thing in particular has struck me as the analogy occurred several times:
-
Some experts described the human brain as the most incredibly complex thing that we know about, but they have little idea how it works.
-
Other experts talk about how cleaver machine learning (aka AI) is at doing what it is asked, but they have little idea how internally it performs it’s given task.
From this it would appear that the the two have something in common, in computer science terms they are black boxes. We see information go in and the results come out – but what happens inbetween is a mystery. It strikes me that the technology may be advancing leaps and bounds but our understanding is not keeping pace.