Hyperautomation company Hyperscience Inc. today announced updates to its Hypercell platform that are designed to accelerate automation for a wide variety of back-office documents, use cases and processes.
The new updates are focused on helping organizations further embed machine learning and artificial intelligence into the core of their businesses to accelerate digital transformation efforts. The updates seek to address the issue wherein 70% of digital transformation projects, according to a McKinsey study, fail to achieve their stated goals.
Hyperscience argues that this outcome is due to traditional intelligent document processing, robotic process automation and optical character recognition technologies falling short in delivering on the promise of digital transformation. That’s because they’re typically rigid and rules-based and hence struggle to adapt to new, varied and complex document types inside an organization.
Hypercell, built with AI at the core, delivers accuracy rates of 99.5% and automation rates of 98%, with modular blocks and workflows that deliver seamless process orchestration and integration with downstream enterprise systems, according to the company. The new version of the platform enhances the platform’s existing capabilities with new innovations in models, model management, workflow orchestration and cloud and on-premises infrastructure. The goal is to prepare organizations in persuing new initiatives that unlock generative AI applications for mission-critical applications inside the enterprise.
The update today introduces a deep learning model that automates long-form document extraction, enabling organizations to process complex documents such as contracts and loan applications more efficiently. By extracting critical data points and understand nuanced content, the system reduces the need for costly experts, streamlines operations and enhances decision-making.
The update also improves AI and machine learning model lifecycle management by offering seamless upgrades and enhanced model portability. With the service, organizations can now adapt to new document types without extensive retraining and, in the process, preserve their existing investments in previously trained models while accelerating the adoption of new features.
Additionally, the new release offers enhanced governance and security with features that ensure compliance and transparency. The system’s automated audit logs provide clear traceability, making it easier to maintain regulatory standards and enabling organizations to integrate these AI-driven processes with confidence.
“Digital transformation starts with data and requires the accurate processing and automation of all the information assets that flow through an organization,” said Chief Executive Andrew Joiner. “This transformation has eluded most companies who have had to rely on legacy technologies and business process outsourcers.”
Joiner spoke with theCUBE, SiliconANGLE Media’s livestreaming studio, in March, when he discussed Hyperscience’s vision for AI-driven enterprise efficiency, including how at the heart of the company’s approach is the ability to process “human-friendly information,” unstructured data that defies conventional database storage: