ABBYY adds 22 patents to expand AI document automation

ABBYY has secured 22 new patents over the past two years, expanding its portfolio to more than 400 patents and patent applications worldwide. The company continues to focus on software that processes documents and business data.

The newly issued patents cover language detection, user interface design, image identification, information extraction and document decoding. Several relate to optical character recognition (OCR) and computer vision, core technologies in document automation.

Document processing vendors have increased their focus on intellectual property as more organisations invest in automation and experiment with generative AI tools. The sector has also faced growing scrutiny over accuracy, auditability and controls when AI systems handle business records.

Patent focus

ABBYY's latest additions include patents titled "Extracting Multiple Documents from a Single Image" and "Detecting Fields in Document Images". The topics reflect common enterprise use cases, such as processing mixed batches of scans, identifying form fields and separating individual documents from a single capture.

Other patents address information extraction and machine learning methods for document understanding, including "Identification of Key-Value Associations in Documents Using Neural Networks" and "Continuous Learning for Document Processing and Analysis". These approaches often underpin how systems find and link items such as names, dates, invoice totals and reference numbers within structured and semi-structured documents.

The portfolio also extends into user interface design, including a patent titled "Display Panel or Portion Thereof with a Graphical User Interface". Document automation suppliers often compete on how analysts and business users review extracted data, validate exceptions and manage workflow queues.

ABBYY also highlighted patents related to image and barcode recognition, including "Decoding of Two-Dimensional Barcodes Under Unfavorable Conditions" and "Assessment of Image Quality for Optical Character Recognition Using Machine Learning". Barcodes and variable image quality remain operational challenges in sectors such as logistics, healthcare and manufacturing, where documents and labels may be captured on mobile devices or in difficult lighting.

GenAI questions

ABBYY positioned the patents as part of its broader work on AI for documents and process automation, as enterprises test generative AI for knowledge work. Many companies are evaluating how large language models fit with established automation stacks that already include OCR, classification and workflow tools.

Within document processing, generative AI has prompted debates about reliability and governance, particularly when outputs feed financial, regulatory or operational systems. Suppliers have responded by emphasising higher-precision extraction, validation steps and methods to monitor changes in model behaviour over time.

ABBYY said its work focuses on AI designed for documents and business processes rather than general-purpose models, and that the approach is intended to meet enterprise requirements such as accuracy and compliance. The announcement did not disclose where the patents were issued or whether they were granted in the US, Europe or other markets.

Company background

ABBYY is best known for software that converts documents into usable data, including OCR and intelligent document processing tools. It also sells products that analyse business processes. ABBYY is headquartered in Austin, Texas, and has offices in 13 countries.

More than 10,000 enterprises use its products, including many Fortune 500 companies, according to the company. ABBYY has operated for 35 years and has historically sold document capture and recognition tools to large organisations with high volumes of paperwork and scanned content.

Patent portfolios can serve as both a defensive measure and a commercial asset for technology vendors, especially in competitive categories where multiple suppliers offer similar features. In document automation, differentiation often depends on accuracy rates, support for varied layouts and languages, and the ability to handle edge cases such as poor scans, complex tables and mixed document sets.

The 22 new patents were developed over a five-year period and issued in the past two years, ABBYY said. The company linked the milestone to its product direction as enterprises adopt more automation alongside generative AI.

Ulf Persson, ABBYY's chief executive officer, described the patent progress as part of the company's broader innovation agenda.

"ABBYY's latest patents are a testament to our ability to drive the next level of innovation. By combining cutting-edge AI with a deep understanding of intelligent document processing and business needs, we are shaping the future of intelligent automation and delivering solutions that empower businesses to thrive in an ever-evolving landscape," Persson said.

ABBYY plans to showcase its products at its Ascend events in Nashville and Brussels this spring.