Hyperscience Enables Transformational AI at Scale with Significant Updates to Core Platform

Latest Version of Hyperscience Hypercell Deepens Automation Capabilities and Further Embeds Machine Learning into the Core of the Enterprise

New Deep Learning Models for Complex, Long Form Documents, Industry Leading Model Lifecycle Management Features and Extensive Developer Tools that Accelerate End-to-End Automation

Strengthened Auditability and Transparency Features Maps Hypercell’s IT Governance and Security to the Highest Standards, Helping Organizations Safely Unlock AI Potential

Hyperscience, a market leader in hyperautomation, today announced comprehensive updates to its core platform, the Hyperscience Hypercell, designed to accelerate automation for a wide variety of back office documents, use cases, and processes. These updates help organizations further embed machine learning and AI into the core of their business, accelerate digital transformation efforts, and advance Hyperscience’s mission to deliver trusted, cutting-edge AI capabilities with proven ROI.

Also Read: The Promises, Pitfalls & Personalization of AI in Healthcare

“Organizations today face relentless pressure to modernize in order to reduce costs, improve productivity, and deliver exceptional customer, employee, and citizen experiences”

“Organizations today face relentless pressure to modernize in order to reduce costs, improve productivity, and deliver exceptional customer, employee, and citizen experiences,” said Andrew Joiner, CEO of Hyperscience. “Digital transformation starts with data, and requires the accurate processing and automation of all the information assets that flow through an organization. This transformation has eluded most companies who have had to rely on legacy technologies and business process outsourcers. The latest release of the Hyperscience Hypercell delivers on the promise of transformation by setting the technology foundation to harness the power of GenAI and LLMs, and enabling our customers to convert back office documents and processes into strategic advantage.”

According to a study by McKinsey, 70% of digital transformation projects failed to achieve their stated goals¹. Traditional Intelligent Document Processing (IDP), Robotic Process Automation (RPA), and Optical Character Recognition (OCR) technologies have fallen short in delivering on the promise of digital transformation, since these offerings are rigid and rules-based, and struggle to adapt to new, varied, and complex document types inside an organization. As a result, organizations have been forced to rely on costly, manual effort from Business Process Outsourcers (BPOs) to process document-based workflows inside an organization.

Hyperscience delivers a novel solution to this challenge with the Hyperscience Hypercell. Built with AI at the core and based on proprietary machine learning models, Hyperscience delivers remarkable accuracy rates of 99.5% and automation rates of 98%. Its user-friendly design allows business professionals to train and manage models utilizing their own data and expertise. The platform’s modular blocks and workflows facilitate seamless process orchestration and integration with downstream enterprise systems. Additionally, the Hypercell can be deployed in the customer’s technology environment of choice, and is purpose-built to address stringent accountability, data security, compliance, legal, regulatory, and privacy concerns.

The new version of the Hyperscience Hypercell platform, based on software version R40, builds on these unique capabilities through new innovations in models, model management, workflow orchestration, and cloud and on-premises infrastructure. Together, the innovations prepare organizations to pursue new initiatives that unlock GenAI applications for mission critical applications inside the enterprise.

The new updates to the Hyperscience Hypercell, R40 software version include:

Unlock Critical Insights from Dense Content — New Long-Form Extraction Model Enhances Understanding of Complex Unstructured Documents

Long-form documents, like contracts, insurance policies, credit agreements, stock purchase agreements, and loan applications, are the lifeblood of key processes within organizations. Traditionally, unlocking critical details from these documents required costly experts who could interpret the nuanced context, spot connections between related information, and extract insights buried within varied formats.

The latest version of the Hypercell provides new capabilities to its deep learning model that enable Long Form Extraction and automates this understanding, allowing enterprises to streamline decisions and operations by extracting critical, interdependent data points and multiple occurrences from the complex, unstructured content. By understanding the subtleties embedded within long-form documents, customers can now pursue more advanced use cases that enhance existing processes through the ability to efficiently summarize content, understand context, and execute advanced NLP tasks directly within Hyperscience.

Comprehending these details accurately reduces manual review time, accelerates decisioning, and improves customer service. Furthermore, once this nuanced and unstructured content has become machine readable, it can be leveraged to train LLMs in the language of an organization’s business, to deliver more relevant and precise GenAI experiences. Hypercell for GenAI automatically annotates, labels, and structures data from documents for fine-tuning LLMs and GenAI experiences, allowing organizations to rapidly and continuously develop highly accurate and valuable enterprise models.

Also Read: AiThority Interview with Adolfo Hernández, Technology Managing Director for Telefónica at IBM

Streamline AI / ML Model Lifecycle Management with Seamless Upgrades, Enhanced Model Portability, and Adaptive Handling of Document Variability

As AI and machine learning systems become increasingly integral to daily business operations, effective model lifecycle management is crucial for maintaining seamless execution and compliance with regulatory standards. The complexity of managing model upgrades, governance, and ongoing adaptability often results in resource-intensive processes that slow down innovation and disrupt business workflows.

Hyperscience addresses these challenges with new features that help organizations preserve their investment in previously trained models by enabling seamless model portability across different versions of the Hyperscience Hypercell platform. This Extended Model Compatibility allows businesses to eliminate retraining costs, accelerate the adoption of new features, and maintain continuity in their AI operations.

A common hurdle in model lifecycle management arises when previously unseen document types enter the workflow, causing drift that disrupts established processes. This can lead to delays, errors, and complex interventions, especially when a system lacks the flexibility to quickly adapt to new content variations.

Hyperscience’s innovative Document Drift Management model tackles this issue by enabling business users to follow an intuitive step-by-step guided workflow to review unexpected document types and adjust the system to handle them the next time they are seen. For new customers, this means they do not need to understand all of the various document types they will be processing in production or be required to manually pre-sort them before they bring them into Hyperscience. Whether implementing the Hypercell for the first time or facing a surge in new document types, businesses and systems integrators can leverage Document Drift Management to minimize manual processes and ensure consistent, large-scale document processing — ultimately reducing operational costs and boosting efficiency.

Also Read: Data Monetization With IBM For Your Financial Benefits

Strengthen IT Governance and Security through Comprehensive and Enhanced Audit Logs to Increase Trustworthiness and Transparency for an AI-First Future

As organizations increasingly leverage the transformative power of AI, ensuring robust auditability, transparency, and governance becomes essential to build trust, mitigate risks, and comply with evolving regulatory standards. In document-centric workflows, auditability provides clear visibility into every decision, enabling organizations to trace actions, validate outcomes, and quickly address discrepancies. This level of transparency strengthens governance practices and ensures consistent compliance, even as AI-driven processes evolve.

With the latest release of the Hyperscience Hypercell, organizations have access to enhanced audit logging capabilities that will help customers understand everything that happened in Hyperscience – who performed the action (the machine or a human), when, and where the action took place. This helps ensure accountability, accuracy, and compliance, thereby enhancing operational transparency and efficiency related to all workflows created within Hyperscience.

This capability guarantees the highest level of security and compliance for handling sensitive government data, ensuring that all actions within a Hyperscience workflow are both traceable and secure, which are critical in Hyperscience’s pursuit of FedRAMP High authorization by the end of 2024.

The new updates to the Hyperscience Hypercell are now available to all customers upon upgrading to R40. These updates are a significant advancement in automation technology, offering enhanced capabilities that understand complex and unstructured documents, streamline AI/ML model lifecycle management, and strengthen IT Governance and Security for enterprise-grade deployments that will deepen end-to-end process automation for organizations around the world.