The adoption of AI by advisers and financial businesses is happening organically, according to Jacobsen.
“Individuals and teams are actively looking for opportunities to improve the way they do things,” he says.
“It’s a big change from when we had to stand up large IT teams with complex change programs. I think advisers and their teams are a lot more sophisticated when it comes to technology and are rapidly adapting their business models.”
He adds that the advice industry is currently unable to service the volume of demand for advice, and “certainly will no longer be servicing the mass market population”.
“AI is a tool that enables advice to be delivered more broadly again and be more accessible to more people,” he says.
“That’s a good thing for society and advisers will continue to thrive in that future, able to reach more clients, and add more value.”
Although there may be the natural anxious reaction from advisers that AI or other technology could replace them, research from Neilson & Co Wealth managing partner Ben Neilson has asserted that AI tools will, at best, complement industry professionals rather than replace them.
US President Joe Biden signed an executive order on Monday 30 October regarding AI. It aims to find a middle ground between the interests of high-tech companies, national security, and consumer rights, and will establish some initial rules that can be strengthened through laws and international agreements.
President Biden said AI is driving change at “warp speed”.
“To realise the promise of AI and avoid the risk, we need to govern this technology.”
Data quality central to effective AI
Ensombl CEO Clayton Daniel acknowledges that the financial sector initially hailed ChatGPT as a breakthrough; however, it has a crucial limitation.
“It’s a standalone product,” he explains.
“It’s not really plugged into anything else.”
The heart of AI’s effectiveness lies in the quality of the data it processes. Daniel emphasises the adage “quality in, quality out,” indicating that well-structured, accurate, and clean databases are essential.
A recent report from Netwealth has warned against the risks of putting client information into public third-party systems like ChatGPT.
“From a financial planning point of view, you can’t put personal client details into ChatGPT,” Daniel says.
“However, if you are using something like PrivateGPT, that information is not going elsewhere. You’ve got really good data controls, cleansed data. Then you go on to train your artificial intelligence within the PrivateGPT environment.
Daniel adds that AI can be further trained in the PrivateGPT environment.
“You can deliver quite a bit of value to your own internal company in terms of efficiencies,” Daniel says.
“You wouldn’t use it to come up with suggestions on advice at this stage. You wouldn’t use it to recommend financial planning strategies.”
PrivateGPT is a fully functional AI product that lets individuals ask questions about documents using powerful large language models (LLMs), even without an Internet connection.
Daniel adds that there are some standalone paraplanning AI products coming to market that are designed to assist with the paraplanning process.
“I see AI, combined with QAR [the Quality of Advice Review], as reduction in reporting requirements, ultimately, more efficiencies within the production of advice,” Daniel says.
“The best news about all of that is the fact that advisers will be able to see more clients more regularly and paraplanners will be implementing more work and speaking to more clients. So, all good things as far as I’m concerned.”
paraplanner.ai chief growth officer Alex Gassner agrees.
“I think the paraplanner’s role is probably going to morph into being much more involved with clients and client meetings,” he says.
“Fifty to 70 per cent of the [paraplanning] role will be able to be automated slash be taken care of by the AI agent model.”
Gassner adds that many financial planning firms are “probably” using large AI language models to aid in marketing efforts or to create meeting and call summaries for clients.
“I think there needs to be more clarity [around AI] from a regulation point of view,” he says.
“Where is the data? Where does it go? Who has access to it? Are the large language models cross-trained with the data or are they not?”
Gassner expects this need for clarity to become more evident within the next 12 months.
“I think that that just gives confidence, and that’s what businesses need right now. Confidence that they’re going to invest in this technology that absolutely does work at the moment and will only get better.”
Gassner, along with his business partners Denis Konoplev and Fauzi Fellal, recently debuted paraplanner.ai in the US market. Its aim is to automate time-consuming tasks and give frequent, personalised attention to financial planners’ clients through AI.