In this article I will explain the difference by citing a simple example , rather go by definitions.
Consider yourself to be a manager in the credit card division of a bank. You have been given an assignment to reduce the customer credit default by 10% in Q3 -2019. The bank has an IT department which has data science team and ML engineers.
If I had been the manager of the bank , following would have been my approach.
Step 1: Collect all the data points of our credit card customers , in terms name , locality , profession , age , marital status , income and lot many customer data points , something similar to the below table and yes there can be loads of more data.
Step 2: I would have asked my Data Science team to generate insights from the data , in terms of clusters which are performing good , identifying segments which are performing bad , what are the reasons why customers are defaulting and all. My expectation would be to come up with a deck of all the insights.
Step 3: After having understood the insights, understanding multiple parameters. I would have asked my Machine Learning engineer to develop a model which could help the sales team analyse whether to give a credit card to the new customer or not , on the basis of customer inputs.
Now you can see , the work of data scientist is to explore the data for insights and deliver a report , where as the work of Machine Learning is develop a model/software which can be used to take decisions.
Yes , both are a part of A.I , as they both use A.I tools to deliver to results.
I hope i was able to simplify and not create more confusion.
You can write to me on email@example.com or please leave a comment below.