WealthArc launches AI agent to structure portfolio data

WealthArc has launched an AI Agent that converts financial documents into structured investment data for wealth managers handling portfolio information from multiple sources.

The tool processes account statements, portfolio reports, private-market investment documents, capital account statements, and performance reports. It can interpret files received as PDFs, spreadsheets and other structured or semi-structured formats, then add the extracted information to WealthArc's wider data service.

The launch targets a persistent operational issue in wealth management: firms often receive portfolio and alternative investment data in inconsistent formats that are not immediately usable for reporting or analysis. Teams frequently have to manually extract, reformat, and reconcile the information before it can be loaded into portfolio systems.

The AI Agent is designed to close that gap by interpreting documents in context rather than relying solely on fixed templates. It operates within a human-in-the-loop model, with users reviewing suggested interpretations and corrections before they are applied to portfolio datasets.

That approach reflects the challenge of working with financial records from different custodians and reporting providers, where terms, structures and assumptions can vary sharply. Even similar statements may present the same holdings, transactions or valuations in different ways.

Artur Kluz, Chief Executive Officer of WealthArc, linked the launch to wider adoption of artificial intelligence across the sector.

"AI won't replace advisers, rather equip them with deep knowledge and flexible workflows. The real barrier to AI adoption in wealth management isn't the models themselves but turning fragmented financial data into reliable, structured data," Kluz said.

Data layer

The AI Agent feeds into WealthArc's broader data infrastructure service, which standardises portfolio information across custodians, asset classes and currencies. Once extracted, the data can be used for reporting, analytics and other AI-based workflows within client systems.

At the centre of that offering is WealthArc Data Box, described as a multi-custodian, multi-currency data engine. The platform aggregates, reconciles and standardises portfolio data for banks, investment managers and family offices.

WealthArc is building what it describes as a global grid of more than 1,000 next-generation data feeds. The AI Agent is presented as part of that broader effort to create a common data layer for wealth and asset management operations.

The launch comes as many firms in private banking and wealth management seek practical uses of AI in back-office and client servicing tasks, rather than only in front-end analytics. A recurring obstacle has been the quality and consistency of source data, especially when portfolios span multiple custodians, currencies, and reporting systems.

Document variation

Radomir Mastalerz, Chief Technology Officer of WealthArc, said the technology was designed for documents that do not follow predictable layouts. "Traditional automation works well when documents follow predictable formats, but financial reporting rarely does. Statements from different custodians often use different terminology, structures and assumptions. WealthArc's AI Agent helps interpret that context and flag potential inconsistencies, while keeping humans in control of the final outcome," he said.

The tool can highlight inconsistencies for review before records are updated. That process is intended to reduce the risk of errors in environments where portfolio data is used for investment reporting, operational oversight and client communications.

WealthArc operates from Switzerland and serves clients across Europe, while expanding in the United States, Asia and the Middle East. It provides data processing and monitoring services intended to turn fragmented custodian records into standardised portfolio datasets for wealth managers, private banks, family offices and financial technology providers.

Kluz also outlined the scale of the company's broader infrastructure build-out. "Our goal is to build more than 1,000 next-generation global data feeds in the coming years. By expanding this infrastructure, we are building the foundational data layer for the wealth and asset management ecosystem-enabling institutions to scale operations, automate workflows and adopt AI with confidence," he said.