Abbove Connector Opens AI Wealth Planning to Advisers

Abbove, the Brussels wealth planning platform, has launched a Model Context Protocol (MCP) connector that lets advisers point their institution's AI tools at structured client wealth data. The move brings AI wealth planning closer to daily practice: an adviser can now ask a question in plain language and have the model read across an entire client book, returning, for example, which families have cash coming free in the next 12 months, with dates and amounts.
The connector runs on the open MCP standard, which Anthropic introduced in November 2024 and which has since been adopted across the major AI vendors. Advisers can connect the tools their institution already runs, such as Claude, ChatGPT, Mistral or Microsoft Copilot, within their existing access rights. Abbove activates the connector per organisation, and it is available from launch.
What can advisers actually do with it?
The connector targets three jobs advisers do most weeks. Abbove says it can rebuild a complete wealth overview from meeting notes and documents for the adviser to review and import, compressing what it describes as days of work into under an hour. It can assemble meeting preparation, an agenda, interview guide and a read on risks and opportunities, in moments. And it lets advisers interrogate their whole client book in natural language to surface individual needs and advisory openings.
Each of these rests on the same premise: the answer is only as good as the data the model can read.
Why does the data layer matter more than the model?
Because the model is only as useful as the data beneath it, and that is where most firms are stuck. Independent surveys through 2026 put AI adoption in wealth and asset management close to universal, yet far fewer firms have pushed it beyond pilots. In the 2026 Asset Managers Report compiled by Funds Europe, 99% of firms reported some AI usage, but 64% said AI will deliver real value only once they improve the quality and structure of their underlying data. Among UK advice firms, intelliflo's 2026 survey recorded AI use rising from 43% to 74% in a year, concentrated on administrative work rather than advice itself.
The constraint is consistent across the research: fragmented, unstructured data rather than the models. A global asset management survey by Grant Thornton found more than half of firms citing limited access to quality data as a major barrier, with about half yet to build processes to clean, normalise and tag it. Europe also continues to trail the United States and Asia on adoption pace.
Abbove's pitch is that it sits on exactly the part of the stack those figures point to. The platform models each family's wealth in a structured form, covering goals, life events, gifts, ownership and family structure, and is built with the client and the adviser rather than aggregated after the fact. The company positions this structured record as the foundation an LLM needs to be useful in advice. Chief executive Guillaume Desclée framed poor data as the limiting factor on any AI deployment in wealth advisory, likening a model without quality data to a Formula 1 car running on bad fuel.
How does this fit Abbove's wider AI push?
The connector extends a run of AI moves rather than starting one. Founded in Brussels in 2017, Abbove earlier introduced an AI agent, Mia, aimed at advisers and family officers. On the company's figures, the platform is now used by more than 1,450 advisers serving over 40,000 families, representing more than 200,000 individuals across Europe. Its white-label model has been deployed inside large institutions, including ING Belgium, which built its ING Financial Compass wealth planning experience on Abbove technology earlier in 2026.
The connector also lands as MCP spreads through financial services, with brokerages and core banking providers opening their own servers to AI assistants over the past year. Abbove's angle is specific to its niche. Where brokerages have exposed trading accounts and banks their core systems, Abbove exposes the structured family wealth record that planning conversations actually turn on.
Why This Matters to FinanceX Readers
For wealth managers and the investors backing wealthtech, the signal is where the value is accruing. Near-universal AI adoption has not translated into advice-grade output, and the surveys keep naming the same bottleneck: structured, permissioned data. Platforms that own the clean data layer, the system of record for a family's full wealth picture, are positioned to capture that value as advisers connect whichever models their institutions standardise on. The connector approach also lowers switching costs for the AI tool itself, which shifts durable advantage towards whoever holds the data rather than whoever supplies the model.
For European firms weighing where to spend, the practical question is whether their client data is structured enough to be worth querying at all.



