I have recently came across a demo of for a software that's built on natural language processing and its job is to convert a text the user types on a Python interpreter such as Jupyter and the software automatically converts it to a code. A detailed explanation of this software can be found here.
I was wondering if such software is developed to handle cleaning and manipulating data, then what would be left for data scientist to do? At the moment there is a trend, the democratization of machine learning, or the citizen data scientist. While more and more software integrates machine learning would this be considered a dangerous trend? and how should we accommodate with the rising trend of similar softwares?