Anthropic’s latest AI release for wealth managers reinforces a growing reality across the industry: firms are rapidly adopting AI, often before establishing proper control over the data foundations beneath it. 

For wealth and asset managers, the potential applications for AI are compelling. AI can accelerate meeting preparation, review earnings releases against portfolio exposures, assist with compliance workflows, generate client-ready commentary, and help relationship managers respond faster and more intelligently to client questions.

Alongside this are a myriad of options of how to embed AI in day to day business operations – large AI companies such as Anthropic, specialist start up providers, DIY self-build tools or working with specialist providers and consultants.

Used properly, these tools could materially improve efficiency across investment management businesses, allowing teams to spend less time assembling information and more time applying judgement, managing risk and strengthening client relationships.

But amid the excitement, there is a risk that firms misunderstand where the real challenge, and therefore the real competitive advantage, resides.

The data.

Where AI and operations collide
AI agents do not arrive preloaded with proprietary investment data, client records, operational workflows or historical and business context. They rely entirely on the quality, structure and accessibility of the information they are connected to, and in much of the wealth management industry, that information still sits fragmented across custodians, spreadsheets, reporting tools, CRMs and legacy systems.

This is where much of the current AI narrative begins to collide with operational reality.

Increasingly, the more thoughtful commentary emerging across the market is recognising that large language models are excellent at synthesising, summarising and interacting with information, but they are not inherently designed to guarantee the accuracy, lineage or governance of the underlying data. That distinction matters enormously in investment management.

An AI-generated insight is only useful if the portfolio data underneath it has been reconciled and analysed correctly. A client report is only valuable if the numbers are accurate and auditable. A compliance workflow only works if the source data is complete, governed and explainable. Without that foundation, firms simply automate confusion.

This is why it’s not the first time, nor the last, you’ll hear “garbage in, garbage out” in relation to the era of AI agents. The fact is, AI amplifies poor operating models rather than fixing them. Fragmented systems, duplicated data, inconsistent calculations and manual reconciliation processes become more visible - and more problematic - once autonomous agents begin interacting with them at scale.

The firms that succeed with AI are therefore unlikely to be the firms that merely adopt the latest tools fastest. They will be the firms that own and control their data architecture and become truly data-first businesses.

Modern investment operating models
Historically, many wealth managers built technology stacks application by application, adding reporting tools, CRMs, portfolio systems and client portals over time. The result was often a collection of disconnected technologies, each maintaining its own version of the truth.

AI exposes the weakness in that approach almost immediately. They need consistent investment data, standardised calculations and reliable historical records. They require context and memory. Without that, even sophisticated AI agents struggle to produce outputs that investment firms can genuinely trust.

This is precisely why Point focuses on Investment Data Intelligence.

The Point Investment Data Intelligence (IDI) platform was designed to provide wealth and asset managers with the proprietary, AI-ready data orchestration foundation modern operating models now require. By aggregating, reconciling and structuring investment data into a unified Independent Investment Book of Record (IIBOR), Point enables firms to create a reliable and transparent data layer across their businesses.

That foundation then supports analytics, reporting, workflow automation and, increasingly, AI applications - safely and consistently.

Importantly, this is not about replacing people with AI. The real opportunity is enabling investment professionals to operate with better intelligence, better context and faster access to trusted information.

AI agents are accelerating rapidly, that much is clear, but as the market moves beyond experimentation and toward enterprise deployment, firms are beginning to recognise that owning the data foundation - not simply accessing the AI model - is what will determine who actually captures long-term value.

Because in wealth management, intelligence is only as good as the data behind it.

Read the original article here.