DIRK internal knowledge assistant (R+V Versicherung)
by R+V Versicherung · Primary insurer
Market unit: DE
R+V Versicherung uses DIRK ("digitaler R+V Kollege"), an internal generative-AI knowledge assistant built on Kauz.ai's no-code aiSuite platform. Piloted from mid-2023 in the internal service (Innendienst), it answers employees' questions on insurance terms and case handling from SharePoint, intranet, Teams and Outlook content, aiming to reduce internal follow-up queries. The system uses retrieval-augmented generation with combined keyword and vector search, a multi-agent set-up that routes questions by topic and can draft and email texts, and several LLMs chosen per use case. Specialist editors review dialogues and maintain content; R+V reports answer quality of 95%. According to an R+V practitioner report of December 2025, the pilot succeeded and department-specific assistants are being rolled out.
1 free view left this month
Business context
- Value chain steps
- Portfolio administrationInformation technology
- Insurance branches
- Property insuranceIndividual life insurance
- Application scope
- Internal only
- Country or market unit
- DE
Value and impact
- Added value category
- Efficiency gainQuality improvement
Technical details
- AI category
- Generative AINatural language processing
Compliance and governance
- EU AI Act risk category
- Limited risk
- EU AI Act role
- Deployer
- Relevant policies
- EU Artificial Intelligence ActGeneral Data Protection Regulation
Metadata
- Development type
- Minimum viable product
Publications over time
When this solution was reported on. Rumors are unverified mentions.
Community
Average rating
Comments
Log in to join the discussion.
No comments yet. Be the first to comment.
Related solutions
Swiss Re applies AI to review and manage contract wordings across its in-force reinsurance business, led by Group Underwriting's Contracts & Wordings Services and P&C Reinsurance underwriting. An existing Swiss Re application combines keyword search with conversational AI so that underwriters and contract specialists can query large contract volumes, while GenAI-enabled analytics identify wording patterns, accumulations and discrepancies across portfolios. Swiss Re stated in June 2026 that these capabilities are being introduced incrementally into production and are still at an early stage. No metrics are published beyond the baseline that a 2020 COVID-19 wording review of tens of thousands of contracts took about 45 days. Underwriters remain responsible for interpretation and decisions.
Mobi-ChatGPT is Die Mobiliar's data-protection-compliant generative AI assistant for employees, announced on 13 September 2023. It runs on Microsoft Azure OpenAI Service inside Mobiliar's own cloud with a custom frontend; queries do not go to OpenAI's public interface and are not stored. Employees use it to draft, edit, summarise, translate and proofread texts, generate ideas, answer questions about documents and check code; Mobiliar supports it with AI training modules and communities. At the Bern-West general agency (45 employees), where it started as a pilot, many staff had used it for about a year (Microsoft, 2025) for customer correspondence, summarising long e-mail threads and marketing research, reporting time savings and better communication quality without figures. The number of users is not disclosed.
by Wüstenrot Gruppe (Austria)
AT
The Wüstenrot Group, a Central European financial group offering insurance, savings, financing and pension products in Austria, Croatia and Slovakia, has used Mindbreeze InSpire since October 2014 to classify incoming mail automatically. The system handles about 25,000 items a day from paper, email, attachments and web channels, assigns them to 138 document categories, extracts more than 20 metadata fields and routes them to specialist departments; once a set accuracy level is reached, items go through without manual checks. It was trained on about 60,000 pre-classified documents after a multi-vendor proof of concept. Mindbreeze reports 85-95% accuracy in live operation (vendor-reported, 2020 case study); no more recent status was found.