The news that the giant AI technology provider Anthropic launched Claude for Financial Advisors was received with some skepticism by industry analysts, but overall, many wealth management technology providers and consultants shared their excitement and enthusiasm, noting that the company had presented them with something that could, if implemented well, solve many issues with current advisor tech stacks.
“I find it positive that Anthropic has sent these signals to the market and it is good for the industry overall,” said longtime analyst Alois Pirker, who is now the CEO of his own consultancy, Pirker Partners.
“Nonetheless, my sense is that Claude will hit the same problem as Salesforce, getting to the level of specificity that financial advisors need,” he said, referring to the behemoth CRM provider and its initial rollout of Salesforce for Financial Services, which reached general availability a decade ago (and later changed its name to Salesforce Financial Services Cloud).
One way in which the technology industry responded to advisors’ desire for such specificity was the development of the specialized and popular advisor CRM product Practifi, he said, which is built on top of Salesforce and caters to certain types of firms in need of more customization.
He said the level of outreach to industry participants and providers ahead of Anthropic’s launch announcement was a good sign, but that many more specific processes within advisor workflows still need to be covered.
To illustrate his point, Pirker cited LPL Financial and its more than 30,000 advisors.
“Within that one firm you probably have 15 different types of advisors, all expecting or needing very different and specific things,” he said, elaborating on the different workflows supporting the various types of products or services each provides to their clients.
Another longtime industry analyst and consultant also sounded a note of skepticism.
“The Kitces stat Anthropic cited that a typical advisory practice spends only a sixth of its time in client meetings is real, and it’s been real for years—what matters now is whether Claude actually moves that number,” wrote Will Trout, director of securities and investments practices at the research and consulting firm Datos Insights, in an email exchange.
“A lot of vendors have promised workflow efficiency and the connectors to Schwab, Addepar and Orion are the proof of concept, now to see if integration works, [in which case] this becomes a margin story,” he said.
“They’re not positioning this as vendor-versus-advisor, but as a layer that makes the existing stack work better, the fact that both advisory firms and portfolio platform vendors like Zocks are aligned is the signal here,” said Trout, adding that the news is more about whether big firms and large numbers of advisors get there first using what Anthropic has built.
“If State Street, Schwab and Fidelity all plug into Claude in the next 18 months, that becomes the standard integration layer, and whoever is slow to connect starts looking like they’re not serious about advisor productivity.
One such technology provider with a large footprint in the RIA industry and an Anthropic collaborator for Claude for Financial Advisors is Orion.
“The net positive for the industry is more access, less friction, and deeper advice—advisors spend less of the day assembling information across systems and more of it on the client conversations that actually require them, working from data that stays portable, permissioned, and governed by their firm,” wrote Reed Colley, president of Orion Advisor Technology, in an email exchange.
Colley wrote that Pirker was “largely right,” in his assessment, and that is the point behind what Orion had built, both on its own and in its Anthropic connectors.
“We built Denali AI and these connectors [in Claude for Financial Advisors] to solve for it: a general model brings the language, the advisor’s data brings the answer as no large language model is going to know a firm’s book of business, and it was never supposed to because that data lives in the systems advisors run their businesses on, which for thousands of advisory firms are Orion Connect and Redtail,” Colley wrote, noting that securely bringing into those conversations an advisor’s own permissioned Orion data under the entitlements and governance their firm has already set, with each person seeing only what they are entitled to see is key.
Several other providers and consultants were less skeptical about the news but reinforced the need for domain expertise that Pirker raised.
“This is the most important thing to happen to our industry in a decade,” wrote Rafael Loureiro, co-founder and CEO of Wealth.com, in an email exchange, “and the LLM providers are solving the genuinely hard part: reasoning, language and the orchestration of their work.”
He pointed out, though, that a large language model is a reasoning engine, not a domain one, and while it can read a trust beautifully, it will still not know what that trust does to a family’s tax exposure in New York in 2027 because that answer doesn’t live in language.
“It lives in a structured model of the balance sheet, the jurisdictional rules (e.g. the New York State “estate tax cliff”), and a calculation engine that has to be right every single time, not most of the time; the depth advisors need comes from what the model is connected to, not from the model itself, and that’s what Claude for Financial Advisors solves,” he added.
Complex planning, he pointed out, is not primarily a language task; in reality, it’s three tasks stacked together.
This includes extraction (turning a 90-page trust into structured data), computation (deterministic tax and estate math that has to be auditable), and judgment (what to actually do).
LLMs, Loureiro noted, are excellent at the first and the third of those tasks, but the second is where general-purpose tools break because in a regulated business, a number an advisor puts in front of a client has to be reproducible and defensible, which means it comes out of a deterministic engine, not a probabilistic one.
A technology consultant who has been working with advisory firms of all sizes as an AI steward in bridging the gap between existing technology stacks and newer AI tools, including Anthropic’s Claude, among others, also saw that company’s news as a big positive.
“We view this week’s announcement from Anthropic as a material moment in the deployment of AI for financial advisors,” said Mike Shannon, co-founder and CEO of Impruve, “not just because of the thought put into the core offering of being able to take certain domain-specific skills off the shelf and go to work using Claude.
He said that, from a foundational standpoint, the work required across the wealthtech stack to get it to its current state represents a fundamental shift from what he calls a closed or walled-garden posture to an open garden posture.
“We believe it’ll become an imperative to provide data access, such as by way of an MCP server, if you are to be competitive in the wealthtech stack and we’re excited about the offering and the participants in it this week,” said Shannon, noting that he expects the news to set the stage for what the new normal is in terms of being able to access your own data as a wealth management firm.