Intuit®: Official Blog > Innovation > Engineering > Yes, Machine Learning Can Help Make Tax Preparation Less Painful

Yes, Machine Learning Can Help Make Tax Preparation Less Painful

Engineering, Innovation Ashok N. Srivastava, Ph.D., is Senior Vice President and Chief Data Officer at Intuit. He is responsible for setting the vision and direction for large-scale machine learning and artificial intelligence across the enterprise, to help power prosperity across the world.

Artificial intelligence has been discussed in the technology industry for a long time. However, at Intuit, artificial intelligence is more than the latest buzzword – it is a key technology that can truly change customers’ lives.

I believe that Intuit is in a unique position to help with our customers’ taxes through artificial intelligence, because of the unrivaled set of financial and compliance data we can access.

For most people, doing their taxes is one of the biggest financial transactions of the year, and also one of the most complicated. This can cause anxiety, as well as a lack of confidence that they are filing their taxes correctly and getting all the money they deserve.

At Intuit, we continue to bring our AI and machine learning expertise to the tax preparation experience. This started with building and using our own knowledge engine, which has been available to customers since tax-year 2014.

Through machine learning and artificial intelligence, we can learn from millions of customers what are the useful patterns and challenges in taxes.

For example, if you are a florist, we could look at every other florist who has filed with TurboTax, evaluate the deductions used by these virtual peers, and then recommend appropriate approaches for your taxes. You could just say yes or no verbally to these suggestions, and then we could get it done. That’s our vision.

Today, TurboTax uses artificial intelligence to help customers get their maximum refund by recommending whether to go through the time-intensive itemized deduction process or not. This can save users up to 40 percent of tax prep time and significant document retrieval effort. And, for those who do itemize, TurboTax uses data analysis and machine learning to identify and recommend deductions, including obscure ones.

We also apply artificial intelligence technology to product development, to speed up the development process of getting our tax software to market each year with the latest tax codes and maximum accuracy. Artificial intelligence and machine learning allow us to convert the massive and complex 80,000 pages of the U.S. Federal tax code into an application that simplifies the process for our customers.

Even the way we secure your information is powered by artificial intelligence and machine learning – see the below video where I explain some of these efforts at the recent Amazon Web Services (AWS) re:Invent conference:


If you are a data scientist who has an interest in using your expertise to improve the financial lives of millions of customers, we have some really interesting opportunities for you. Please visit our careers page to learn more, and I hope to hear from you soon.




Comments (1) Leave your comment

Leave a Reply

Your email address will not be published.