Less than 5% of advisers have a full AI policy

For providers, offering advisers practical support with artificial intelligence (AI) in the form of governance packs, real integrations, evidence of value will be key in keeping business, NextWealth has said.

The latest report in NextWealth’s Adviser Tech Stack series – Clearing the path for AI in advice – ranked the material barriers that firms must address to move from pilots to scale when it comes to AI, with internal governance frameworks coming high up that list.

The findings showed that fewer than 5% of financial advisers have full AI policy, guidelines, and sign-off.

Conducted among 200 advice professionals, the research showed compliance and regulation are the standout barriers to AI adoption, alongside a lack of systems integration, and the findings show that 79% of all firms surveyed believe it poses a barrier to adoption, 19% of which view it as a very significant barrier. This rises to 23% for sole traders.

NextWealth report director and associate research director Chanelle Paynter said many of the foundations of AI “aren’t in place”.

She said: “Firms have rightly chased the quick wins, trailing meeting note-takers and similar tools, taking a tactical approach to adoption.

“Promises that AI will ultimately help to ‘close the advice gap’ by doubling capacity and cutting the cost-to-serve, will be nigh on impossible to realise without core building blocks such as clear governance and policies, reliable data connectivity and integrations, consistent processes and a simple strategy.”

Paynter added that providers have a “clear role” to help financial advice firms with tackling AI compliance, and it is “in their own interests to act”.

“In a market where more than half of firms are open to changing AI provider within the next year, practical support in the form of governance packs, real integrations, evidence of value will be key in keeping business,” she said.

Need for rigorous due diligence ‘isn’t abstract, it’s practical’

NextWealth’s report highlighted that the concern advisers express about the need for rigorous due diligence “isn’t abstract, it’s practical”.

A lack of a structured due diligence framework was cited by 77% of financial advisers as a barrier to AI adoption in some form.

“Firms want proof, not promises. They want plain English explanations of how these models work, where data sits, how it’s tested, and how client detriment is avoided,” said Paynter. “And they worry that they may not yet know what the right questions to ask are. They also worry whether providers can provide straight, audit-ready answers to those questions once known.”

NextWealth set out that firms have proven that they are able to trial, learn and adopt low-risk, repeatable use cases of AI that deliver “visible gains in days, not months”.

However, Paynter said few firms are tackling the bigger changes that could “reimagine the advice process”.

Paynter says suitability assessment and reporting provide the standout opportunity, but progress is slow for two reasons: many tools still fall short on accuracy and logic, meanwhile AI amplifies the process you already have. If templates and checks are inconsistent, AI will scale the inconsistency, so firms are pausing until these issues have been addressed.

Return on investment questions are also surfacing, NextWealth’s findings showed.

“Firms are weighing license costs against time saved, especially where AI sits alongside existing templated processes and still relies on manual data entry. AI has proven its efficiency gains, but its long-term impact will depend on solving integration and data quality issues,” said Paynter.

NextWealth’s research found that where AI has been adopted, the benefits being felt are well documented. This includes increased speed and operational efficiency, which are cited as key gains by 84% of advice professionals who took part in the research.

However, less well documented is the fact that many firms do not yet have a long-term plan for reinvesting the time AI frees up.

“It is not just AI governance and due diligence gaps we need to be mindful of, it seems there is a post-AI adoption strategy gap emerging too,” concluded Paynter.