Soon, perhaps as early as next year’s tax season, a major milestone will be officially reached with the goal of fully digitizing and automating federal tax returns, which will significantly reduce processing times and, in ultimately, the delivery of tax refunds. But starting today, San Francisco Bay Area robotic scanning company Ripcord announced that it has entered the next stage of a new pilot program to digitize archived tax returns as part of an IRS technology pilot program, improving the IRS’ use of data to reduce the manual process of entering and verifying tax forms and reliance on paper returns.
Under the IRS Pilot Solution Challengein collaboration with the IRS Business Digitization and Case Management Office (EDCMO)Ripcord said it has moved into phase two, where it will complete a six-month program to safely digitize taxes Form 709which captures transfers of assets that may be subject to federal gift tax and certain generation-skipping transfer taxes.
“This is a small step toward the giant leap of making the IRS fully digital,” Ripcord President and CEO Sam Fahmy said in an interview with ZDNet. “This year we’re making sure everything is optimized, everything is working seamlessly through a multi-phase process with the IRS. So I have a feeling it will be fully operational next tax season,” did he declare.
Over the years, the IRS has made strides in using digital technologies to streamline the tax filing process, particularly in its e-File program, but technology to properly scan, process, and archive tax forms does not exist. Ripcord uses a combination of machine learning and robotics in its drive to digitize tax returns. “You need advanced machine learning, which is the ability to fully understand the context of the data,” says Fahmy. What does it mean? Let’s say you are extracting data from a tax form. To be able to verify, validate and classify the data requires the use of machine learning technology – the same technology used in self-driving cars. “Consider that in the tax world, the equivalent of a road is a tax document, so we built machine learning models that become very good at understanding what’s in the document,” said Fahmy. So, just as a self-driving car knows everything about a road, from curves and bumps to signs and signals, the same concept applies to a tax document. Each time the machine learning model sees the document again, it understands the amount entered in a specific field and can check it against other similar documents and be able to automate this document.
Also: Best Free Tax Software: Free Online Tax Filing
On the other hand, Ripcord uses robots to take paper documents and turn them into digital documents. This includes ensuring stapled attachments are removed from repositories, determining if the document is properly aligned, sizing the document, etc. “It’s a combination of these two technologies that enables end-to-end automation of tax forms,” Fahmy said. “We are able to process paper and digital documents the same way, so if it’s a paper document, the first thing we do is scan it in a very simple way for the next step, which is machine learning mining.”
So what happens to this data? Ripcord has implemented a system to facilitate the retrieval of this data by the IRS. Because new machine learning technologies can complicate and disrupt other technologies in place in an organization as large as the IRS, Ripcord handles all smart mining; IRS technical staff do not need to be trained on Ripcord’s technology, and Ripcord does not need to install its technology on their technology stack; it is totally non-disruptive, according to Fahmy. “We call it ‘modernizing in place’ so they can keep everything in place while modernizing the process without having to change their own systems and people,” he says. Additionally, Ripcord’s system adheres to the highest government safety standards.
Also: Income tax guide for tax season
Lessons learned from digitizing Form 709 will then be applied to other IRS tax forms, to the point where all IRS forms will be digitized and archived using machine learning technology, less in theory. It’s a noble but worthwhile goal. The Covid pandemic has shown that the IRS cannot always rely on people to process forms as backlogs have accumulated when issuing refund checks. And, as more Americans have had to file their tax forms remotely during the pandemic, there is a greater need to automate the filing process.
“It’s a small step; we don’t want to pretend that we’re tasked with completely modernizing what I have left to modernize at the IRS,” Fahmy said, noting that getting the contract to help the IRS was a highly competitive process. “It is a small step that we hope and plan to continue on this path towards full digitization. There have been more initiatives [by the government] looking for new technologies and small businesses like us. We are a small company of 100 to 150 people. It’s fantastic to see that our government has open arms to companies like us,” he said.