It doesn't sound so serious. It's only been measured since the 60's and the alleged shortening was 0.00159 seconds. The quoted article lists 5 possible causes and it seems reasonable that it's likely a combination of factors.
Apparently each day isn't the exact same length so that 0.00159 seconds doesn't mean much without a mean and standard deviation. One data point doesn't establish a trend and it certainly doesn't establish that there's any kind of problem.
Unfortunately the reporting on "science" isn't usually any better than the reporting on anything else...it's pretty poor.