ACORD Transcriber Expands AI-Enabled Features for Intelligent Document Processing

Large language models, optimized for insurance industry, allow creation of structured data from any insurance document.

(Image source: ACORD Solutions Group.)

ACORD Solutions Group (Pearl River, New York), a provider of next-generation digital solutions for the global insurance industry, has announced the launch of extended capabilities in ACORD Transcriber, a widely used tool that enables digitalization and automation of data processing from insurance forms and other documents. These AI-based capabilities facilitate data extraction from both structured and unstructured document formats, including submissions, contracts, invoices, endorsements, loss runs, market reform contracts (MRCs), schedules of value (SOVs), declaration pages, and more, according to an ACORD Solutions Group statement

The latest features of ACORD Transcriber’s insurance-specific AI/ML engine include new and enhanced capabilities such as the following, as described in the statement:

Comprehensive baseline AI models: Built into ACORD Transcriber are pre-trained AI models for a wide variety of insurance document types. Leveraging these in combination with client samples reduces onboarding time and allows users to quickly achieve high levels of data extraction accuracy.

Advanced Large Language Models (LLM) integration: Sophisticated, insurance-specific AI models and advanced Natural Language Processing (NLP) features allow users to extract data from any insurance documents, even the most complex and unstructured ones. ACORD Transcriber includes out-of-the-box support for integration to multiple external LLMs, such as clients’ proprietary or third-party models.

Improved data security: ACORD Transcriber has the ability to run all data extraction and population use cases in memory, with the option to not persist any customer data.

“Insurance stakeholders have to process a flood of documents every day, in a daunting variety of formats, from standardized to completely unstructured – PDFs, spreadsheets, tables, even text buried in the body of an email or images of paper forms captured by a mobile phone,” comments Rakesh Tangri, SVP & Global Chief Solutions Architect, ACORD Solutions Group. “They need a truly intelligent tool to understand all possible formats, extract and populate data with confidence, and make that data consumable – ultimately allowing them to thrive in our industry’s digital future.”

ACORD Solutions Group reports that, to date, (re)insurers, brokers, and other stakeholders across the global insurance industry have processed millions of insurance documents and extracted over a billion data elements using ACORD Transcriber, resulting in faster speed to market, reduced operating costs, and enhanced data consistency and accuracy. ACORD Transcriber can be accessed through a self-service portal, or by API-based integration with a wide variety of third-party and legacy systems.

“One size does not fit all when it comes to insurance document processing,” ACORD Solutions Group’s Tangri adds. “ACORD Transcriber’s powerful baseline AI models can be combined with a user’s own models, and refined by rules-based training tools, to tailor its capabilities to any organization. If any of your processes currently involves the manual rekeying of information, ACORD Transcriber can show you a better way.”

A demonstration of ACORD Transcriber’s data exchange capabilities, integrated with Salesforce’s Financial Services Cloud, will be held at ITC Vegas, November 2nd at 11:50am, on the Expo Hall Demo Stage.

Guidewire Software Selects ACORD Solutions Group’s ADEPT

Anthony R. O’Donnell // Anthony O'Donnell is Executive Editor of Insurance Innovation Reporter. For nearly two decades, he has been an observer and commentator on the use of information technology in the insurance industry, following industry trends and writing about the use of IT across all sectors of the insurance industry. He can be reached at AnthODonnell@IIReporter.com or (503) 936-2803.

Leave a Comment

(required)