Artificial Intelligence is more than ‘the next big thing.’
The potential for computers to outsmart humans is hardwired into our culture and consciousness, and all the indications are that we are living in the final days of our intellectual supremacy.
Estimated in terms of computations per second per $1,000,
computers overtook the mouse brain in 2015, putting them at
about a thousandth of the human level. This doesn’t sound like
much until you remember that we were at around a trillionth in
1985. The current trajectory will see Artificial Intelligence overtake
the human brain affordably by 2025. While the learning and
improvement mechanisms necessary to make that intelligence
general rather than specific still require work – most current
forecasts for superior general artificial intelligence come around
a decade later, around 2035. At this point the speed of AI
improvement will be such that computers will whizz past us,
making us almost infinitely ‘stupider’ in a matter of months.
The major existential issues raised by this critical event in human
history will not be addressed in this paper. Instead, it will focus
on what is commercially achievable with AI as it stands today:
– What are the currently available AI technologies and what
commercial impact are they having?
– What are the potential benefits of these technologies
to businesses based in cloud-connected data centers?
– Are these benefits significant enough to justify a shift
to AI-driven applications?
– If they are, what needs to be done by way of preparation