Loading use case index…
Loading use case index…
AI use case
UNIQA Poland deployed an AI Knowledge Assistant built on the Polish language model PLLuM and a strict RAG pipeline to give agents real-time, verified answers drawn from thousand…
Core facts from this catalog record. Primary narrative lives in the hero above; full raw fields follow in the next section.
Every column from the source row, in stable order. URLs open in a new tab.
Title
UNIQA Poland: AI Knowledge Assistant on PLLuM Delivers Real-Time Verified Answers to Nationwide Agent Network
Content
UNIQA Poland deployed an AI Knowledge Assistant that draws on thousands of insurance documents — including General Insurance Conditions (GIC/OWU), tariffs, and regulations — to give agents real-time, verified answers via the Polish-language model PLLuM and a strict retrieval-augmented generation (RAG) pipeline. "For UNIQA, the implementation of the Intelligent Knowledge Assistant is a strategic step toward modern customer service," said a UNIQA representative. "Through our partnership with RITS, the organization has gained a tool that supports employees in real-time with precise and verified knowledge. This is not just a technological innovation, but above all, a tangible improvement in insurance standards, where the accuracy of information builds customer trust." The deployment was driven by a knowledge-access problem at scale. As one of the largest insurance companies in Poland, UNIQA manages millions of policies across property, motor, and life insurance, with hundreds of product variants and tariffs that change frequently. Before implementation, consultants struggled to search a massive document database quickly, leading to long average-handling-time in the contact center, inconsistent answers across call center, email, and chatbot channels, and lengthy onboarding for new agents. Technology provider RITS structured the rollout as a four-phase roadmap over roughly eight weeks: an audit and discovery phase in week 2, knowledge-base and RAG infrastructure preparation by week 6, model tuning and a pilot with a selected agent group by week 8, then organization-wide scale-up and KPI optimization. The full deployment now covers UNIQA's nationwide network of agents and consultants. Technically, the assistant is built on PLLuM, a Polish-language large model, paired with a strict RAG architecture that RITS positions as providing 100% hallucination control by grounding every response in verified source documents. The system runs on-premise or in a private cloud rather than on public cloud infrastructure, and integrates with UNIQA's CRM and Call Center systems to create a single source of truth across channels. Within the agent interface, the system offers an AI Suggestions Panel that proposes ready-made answers, Automated Summaries that translate complex legal provisions into customer-benefit language, and consistent responses across call center, email, and chatbot. At operational scale, the system is designed to handle massive query traffic from UNIQA's distributed agent and consultant network, drawing on thousands of documents covering General Insurance Conditions, tariffs, procedures, and regulations, and supporting hundreds of product variants across property, motor, and life insurance lines.
Continue exploring AI deployments in the catalog.
Back to use casesCity
Warsaw
Company/Organization
UNIQA
Continent
Europe
Country
Poland
Category
Insurance
Type
Deployment
Id
4861716a-e77d-4dd9-9fbe-63d9544568b7
Created At
2026-06-21T21:58:13.899366+00:00