I'm not even going to bother reading the article. And I apologize ahead of time for my strong views on this topic.
AI is currently having another one of its resurgences both in industry and the public eye. The last time was in the early to mid-90s. But it died out because of a lack of computing resources necessary to drive the algorithms and run the simulations necessary to refine the techniques and bring costs down to make them practical for common usage. The latest phase is being driven by significant increases in computing resources that have served to mitigate many of these limitations.
But AI is fundamentally just a statistical algorithm that improves the performance of a system over time. The output of a calculation is fed back into the input in such a way as to improve the chance of some desired target outcome, such as keeping a car in its lane and avoiding collisions with other vehicles.
Are these algorithms now becoming useful? Absolutely! But in a very narrow scope. So be wary, be constructively critical and, above all, don't be too impressed.