Sapiens International Corporation N.V. has introduced Sapiens Brain, a proprietary artificial intelligence knowledge system developed for its recently announced insurance platform, SapiensAIP. The technology draws on four decades of insurance software development to provide AI agents with specialist knowledge of insurance products, business rules and regulatory requirements.
Sapiens Brain incorporates millions of lines of software code alongside product documentation, technical specifications and design records.
Rather than relying exclusively on information acquired during general AI model training, the system provides a structured insurance knowledge base intended to support more accurate responses and automated decisions.
Paul Wheeler, CEO of Sapiens, said insurers face growing pressure to modernise their technology and operating processes. The company believes combining Sapiens Brain with SapiensAIP offers a different approach to developing AI applications, allowing agents created outside established insurance systems to work with existing software and business processes.
“There’s a unique pressure within our industry to modernise, and AI is the key to unlocking that,” said Paul Wheeler, Chief Executive Officer, Sapiens.
To be a disruptive player in insurance, we have taken AI further than any of our competitors. The unique combination of SapiensAIP and Sapiens Brain allows agents built outside our core systems to behave as though they were born within them – grafted back in like tissue, not bolted on like a robotic limb.
Paul Wheeler, CEO of Sapiens
The architecture separates AI development from the operational systems insurers depend on for daily transactions. Sapiens expects this separation to accelerate product development without compromising the stability required for policy administration and other insurance operations.
Insurance knowledge graph built on 40 years of software development
Sapiens Brain uses an interconnected knowledge graph designed to represent relationships between insurance products, business rules, operational processes and regulatory requirements.
The system resembles an associative memory structure. Information isn’t stored as isolated documents or instructions. Instead, relationships between different elements provide AI models with additional context when processing insurance-related requests.
General-purpose AI models typically acquire insurance knowledge through their training data, including publicly available online material. Sapiens takes a different approach by supplying information accumulated during approximately 40 years of developing and maintaining specialist insurance software.
Its knowledge base includes product configurations, established business logic and exceptions encountered during real insurance operations. Regulatory requirements form another part of this structure.
For example, the version developed for Sapiens Life draws on an extensive history of life insurance products and supporting software. According to the company, reproducing an equivalent collection of specialist information would require substantial development time for a new market participant.
Glen Ceniza, CTO of Sapiens, described Sapiens Brain as an interconnected memory system built specifically for insurance applications. He argued general AI models lack the insurance-specific knowledge structures required for consistent reasoning across complicated operational scenarios. Without access to reliable business information, models risk generating incorrect explanations or unsupported results.
A human brain is a system of interconnected memory, and that’s exactly how we’ve engineered our product. AI models today are extraordinary, but a model by itself doesn’t understand insurance – what it knows, it’s learned from the open internet, with no ontology to reason inside. Without this grounding, a model will drift and hallucinate, however capable it is.
Glen Ceniza, Chief Technology Officer, Sapiens
“Sapiens Brain gives the model the insurance-specific memory that’s grounded, cited and hallucination-resistant by construction enabling it to reason accurately, consistently and with explainability, which is exactly the standard this industry demands”, Glen Ceniza noted.
Sapiens Brain addresses this problem by connecting AI-generated responses with established insurance knowledge and identifiable information sources. The company says this approach improves factual accuracy, supports traceable reasoning and reduces the risk of AI hallucinations.
SapiensAIP uses external AI agents and Model Context Protocol
Sapiens developed its new AI architecture separately from its established insurance software. Traditional insurance platforms prioritise operational reliability, transaction accuracy and controlled software releases. AI development follows a different timetable, with new models, agent functions and technical capabilities appearing far more frequently.
Maintaining separate development environments allows Sapiens to update its AI services without requiring corresponding changes to established insurance applications.
The company connects these environments through Model Context Protocol (MCP), enabling SapiensAIP agents to access functions and information available within its existing software portfolio.
This approach is intended to reduce the manual configuration and interface mapping associated with conventional API connections.
Sapiens also expects to introduce new AI functions more frequently, without depending on the quarterly release schedules commonly associated with established enterprise software.
Wheeler said the company’s objective extends beyond reducing processing times. Sapiens intends to reconsider how insurers handle claims administration, underwriting, system migrations and product configuration.
AI agents would take responsibility for selected operational activities, using insurance-specific information stored within Sapiens Brain. Insurers would retain their established systems while introducing new automated processes through SapiensAIP.
Industry specialists see potential for changes in how insurance work is organised. Karlyn Carnahan, Head of Insurance, North America at Celent, said agentic AI is expected to alter how insurance companies distribute operational responsibilities.
As AI agents assume more coordination, information collection and routine execution tasks, employees would spend more time on activities requiring professional judgement.
Carnahan argued competitive differences would depend less on the number of AI tools deployed and more on how effectively insurers redesign their operations around the technology.
According to Sapiens, combining specialist models with insurance-specific knowledge and detailed activity logs provides a framework for deploying AI across regulated operations.
The company positions Sapiens Brain as the knowledge system supporting these controls, with SapiensAIP providing the environment in which AI agents perform their assigned functions.
Wheeler expects the technology to change how insurers deliver services and manage internal operations. Sapiens sees the launch as an early stage in its broader effort to introduce autonomous AI applications throughout insurance software and business processes.
