Most automation today is NOT “AI”. Most automation uses traditional programming, structured and/or objected oriented, and uses simple data structured and algorithms. What is actually happening is corporate/business taxes are increasing dramatically, inflation is driving all costs through the roof, so to stay above water, businesses have no choice but to automate. AI is over-kill for automation, which is focused on fairly simple, ripetitive tasks. Computer code generators were invented in the late 70s, early 80s because of shortages of programmers, so simple data structures and coding have been automated since then.

So sorry to burst your bubble…it’s the high costs of doing business that’s driving the automation and it has little to do with “AI”.

What is AI?

I hate to spoil the fun, but AI is sophisticated computer programming. Advanced learning and deep learning algorithms essentially are either interpreted or compiled then executed like any other software. Instructions are fetched from memory or cache, executed by various types of processing units sequentially or in parallel depending on the hardware architecture, and then you see some kind of output, which is either execution of a robotic arm, leg, or other body part, painting of graphics through other advanced processors, or just something else depending on the designer’s wishes. Distributed Neural Nets are basically groups of machines with data structures and algorithms mimicking brain architectures.

Bottom line, it all boils down to memory, CPUs, graphics processors, specialized ASICs, FPGAs, and sophisticated software that passes the “Turing Test”.

“AI” remains a marketing term and will be forever more, so please, the hype is nauseating.