Pilot IRS Informational Posting

Pilot IRS - Informational Posting Internal Revenue Service (IRS), Enterprise Digitalization and Case Management Office Digitalization Technology Verticals 1.0 High-Level Summary This posting is part of the Pilot IRS Program. The Internal Revenue Service (IRS), Enterprise Digitalization and Case Management Office (EDCMO) is sharing information about impending Fiscal Year (FY) 2021 procurement needs with its industry partners. Specifically, EDCMO currently intends to issue Pilot IRS solicitations for three technology solutions that support digitalization, called technology verticals: 1) extracting machine-readable data from low-resolution and poor-quality images (digital and paper / hard copy); 2) digital intake and high-speed scanning, including as-a-service approaches; and, 3) Artificial Intelligence (AI), Machine Learning (ML), and Robotic Process Automation (RPA) solutions. The IRS expects this posting to result in three different solicitations in the current fiscal year (FY21), one for each technology vertical, with each solicitation including one or more use cases. It is also important to note that these solicitations will focus on capabilities and outcomes, and not specific tools or approaches. This posting will highlight potential use cases for each of the three technology verticals; these use cases are neither finalized nor the only examples of potential use cases. They do, however, represent the types of challenges that the IRS is facing. This notice will also outline the current thoughts around sequencing for the three technology verticals. This posting is primarily designed to provide information to industry partners so that they are aware of and able to plan (if they so choose) to support the IRS's requirements in this space. We expect an aggressive timeframe for proposals, evaluations, and awards. As a frame of reference, in the previous five instances of Pilot IRS, the approximate duration of solicitation release to contract awards has been roughly 45 calendar days. 2.0 Background / High-Level Priorities At its most fundamental level, IRS is working to: 1) reduce reliance on paper, 2) increase access to machine- readable data, and 3) manage digital data to shift from low-value to high-value work and improve the taxpayer experience. At a near-term level, the first two goals (reduce reliance on paper and increase access to machine-readable data) result in a focus on dramatically increasing the number of IRS forms and correspondence that are digitally available. The increase can be accomplished in three different ways: digital intake, machine-readable data, and digital storage. It is important to note that, far more often than not, these first two goals address existing/legacy images and paper documents, as well as new sets of images and paper that come to or are generated by the IRS during its normal course of business. These first two goals and the corresponding focus on increasing digital intake, machine-readable data, and digital storage are most directly correlated to the first two technology verticals, as set forth below. The third goal (managing digital data to shift from low-value to high-value work and improve the taxpayer experience) includes both improving existing efforts and establishing new efforts to engage with the taxpayer and improve the efficacy and efficiency of IRS operations. This goal frequently does not pertain to existing paper processes; for instance, focusing on providing a different communication channel to taxpayer inquiries via a chatbot, or providing an analysis of cryptocurrency transactions for enforcement purposes, would not focus on existing paper processes. This goal most directly relates to the third vertical, as set forth below Digitalization Technology Verticals With the guidance of and in partnership with the IRS Chief Information Officer, EDCMO will be pursuing potential solutions in three different technology areas (called technology verticals): 1) extracting machine- readable data from low-resolution and poor-quality images (digital and paper / hard copy); 2) digital intake and high-speed scanning, including as-a-service approaches; and, 3) AI, ML, and RPA solutions. This section provides additional context and examples of potential use cases; the use cases are of varying type, scope, and level of specificity. Extracting Machine-Readable Data from Low-Resolution and Poor-Quality Images (Digital and Paper / Hard Copy) Use Case 1: The IRS has a large number of existing digital images that are at a level of resolution (120 dots per inch (DPI) and lower) that creates difficulty in extracting machine-readable data. These low-resolution images reside within multiple systems, and are occasionally available, for a shorter period of time, at higher resolution. Ultimately, some of these images are saved at lower resolution due to the nature of the legacy systems within which they are stored. As much of the information contained within these systems and images is sensitive, the IRS will use less sensitive or publicly available forms/data to confirm the efficacy of a proposed solution before deciding whether to pursue it. Because the images are contained within legacy systems, interoperability with those systems and other IRS requirements (i.e., cybersecurity) will also be a primary determinant of whether a solution will be scaled. Use Case 2: The IRS has a large number of existing documents that are at a level of quality and consistency that it struggles to extract machine-readable data from them. One example of these existing hard-copy images are forms which have retention periods of 50 years or more, and frequently have voluminous attachments. The forms themselves are frequently consistent, but attachments to those forms are of different sizes, different stock, are not consistent in their formatting (placement of data, hand-written and typed), include staples, binder clips, etc., and are at times very brittle. This use case is related to but separate from the high-speed scanning use cases below; in EDCMO terms, this use case relates primarily to digitalization where the other use case focuses on digitization (storing, handling, and transmission of images that may not include machine-readable data). Digital Intake and High-Speed Scanning (Including As-a-Service Approaches) Use Case 1: The IRS has a large number of existing documents which are retained at IRS facilities and the Federal Records Centers. Some of these documents are not accessed on a regular basis, but the IRS must ensure availability of and access to them for enforcement purposes. A primary discriminator of this use case is that it would be focused not just on the actual scanning of documents, but would likely require additional efforts around the managing, distribution, and disposal of physical documents. This use case relates primarily to digitization vice digitalization, as it does not focus on extracting machine-readable data. Use Case 2: The IRS performs intake of myriad hard-copy documents and digital files via existing business processes and systems managed by IRS personnel or industry partners who are under contract. At times, the IRS's interactions with those documents is almost entirely manual, and at other times includes partial digital intake. This includes business processes like correspondence and referrals, which includes unstructured data and information provided by taxpayers. A primary discriminator of this use case is that it would likely require a greater reliance on the analysis of information contained within the images and documents in order to identify the appropriate business process and routing of images/work. Artificial Intelligence / Machine Learning / Robotic Process Automation Use Case 1: The IRS has a significant amount of guidance and publicly-available information which can be accessed by taxpayers 24/7. However, the taxpayer experience may be improved by providing alternate engagement methods, such as chat bots, which would (at least initially) be based on existing and publicly available information, like Frequently Asked Questions and tax form instructions. An additional area of focus could be the use of an interface via a mobile device that would provide information, submission options, language translation, etc. when a phone recognizes an image of correspondence or a tax form. Decisions on whether or not to scale proposed solutions in this space would be supported by real-time data about interactions with users, their likes and dislikes, as well different or additional capabilities they would like to see. Use Case 2: The IRS utilizes publicly-available sources of information regarding cryptocurrency, and an initiative to bring in public, open source data for necessary deconfliction and analytics could demonstrate value to the Service. The process would entail utilizing the data feeds via API calls from current firms with whom the IRS currently has a contractual relationship to determine potential use cases throughout IRS. An automated process would be established to ingest data into an environment for agency-wide access. Open source data could consist of scrapped dark web/clear web information, social media, public records/filings, and technical components such as available Domain Name Registration (DNS), Geolocational IP data, etc. 4.0 Disclaimer This RFI is issued solely for information and planning purposes. It is not requesting comment on the information contained within it, nor is it requesting meetings, documents, or questions at this time. This notice should not be construed as a commitment by the Government for any purpose other than market research. This announcement does not commit the Government to any contractual agreement. The Government is not seeking proposals and will not accept unsolicited proposals. No reimbursement will be made for any costs associated with providing information in response to this announcement or any follow-up information requests. Thank you for your interest. Contracting Office Address: IRS, 1111 Constitution Ave. NW Washington, District of Columbia 20024 Primary Point of Contact: Marcy Almeida contact@example.gov +1 (555) 000-0000