So, how will generative AI impact wealth management?
Some have sighted research from reputable consultancies whilst others have stamped their feet and made lots of noise but the evidence is hard to refute. When I wrote my initial article on AI & ChatGPT at the start of the year it seemed to have come out of nowhere and there was lots of speculation. In the ensuing months more and more firms have made announcements of their developments in the space but this is all something they have been working on in the background for years. This is not a “lets jump on the bandwagon” and start developing, this is the fruition of development that is now being tested and brought under the spotlight for the world to see.
So, the question isn’t “Can Fintech using AI & ChatGPT provide thorough enough financial planning to replace a financial planner?” (as the answer is a clear YES) but “will it replace the human adviser?”
First off there is no black and white answer to this question.
With 27,500 advisers would they all become redundant? It’s very unlikely but if a system can replace them, then would each firm require as many advisers as it currently has? So, with roughly 5,230 advice practices then theoretically you’d only need one adviser per firm – roughly an 80% reduction.
And it’s not just advisers it’s Paraplanners and Administrators too. In fact, the group probably most “at risk” is Paraplanners.
So how do advisers not become at risk? If they turn the other way and pretend it’s not going to happen and therefore don’t embrace technology, they will fall so far behind with their proposition that they are setting themselves up to fail. However, if they embrace technology and utilise its benefits within their business and for their clients then they will be in a much stronger position and stay relevant to client needs.
So, back to the question “Can Fintech using AI & ChatGPT provide thorough financial planning and therefore replace a financial planner?”
Let’s take a deeper dive.
There is no doubt that the use of generative AI / Chat GPT is on the rise. In its first month of release ChatGPT had 57 million users (now at 170M+) and recent numbers show it has in excess of 13 million daily. Those figures are constantly growing.
And to see the onslaught of development in the wealth and advice market you only have to read this month’s AdviceBridge Adviser Insights. Highlights include:
Kasisto launching KAI-GPT, “the world’s first banking-specific” large language model delivering ChatGPT-like conversational experiences, which provides human-like, financially literate interactions at speed and scale.
Orion launched ChatGPT integration for its US clients. It draws upon the previous 20 text messages between the advice firm and the client to interweave sentiment into the dialogue created by ChatGPT, providing knowledge of the advisor-client relationship.
JPMorgan Chase has applied to trademark a product called IndexGPT which will use AI for “analysing and selecting securities tailored to customer needs.”
Range, an all-in-one AI WealthTech platform, announced it had raised $12m in funding led by Google’s AI-focused venture fund.
Firmbase has raised $12 million for its financial planning and analysis platform. What’s interesting is that the funding includes executives from Google, Amazon, and Microsoft.
You only have to re-read the last two to see where BIG tech’s interest lies and what’s coming.
Tech is unlikely to move beyond simple process-driven transactions
There are some who think that technology is for simple tasks such as opening and administering a savings account or Isa. However, AdviceBridge already goes way beyond this with technology that analyses cash, ISAs, GIAs and Pensions making recommendations as to how it should be restructured and way, way more.
Here we’re talking about generative, interactive technology where you can ask it a question, from which it can analyse information – your CIP, for example – and draw conclusions and recommendations.
But technology can’t build trust.
Yes, it can. Maybe not in the same way or as deeply as some may think is required but it certainly can.
Tim Sargisson (ex CEO of Sandringham) supports his argument that clients will only “trust” tech with insignificant sums through referencing surveys that support this. However, these surveys can only ask people about things they don’t know or understand and are therefore flawed. If you’d asked anyone prior to aircraft development whether they’d fly on a metal plane that weighed hundreds of tonnes they would have all said no. Now no one gives it a second thought. Think of the thousands of people who use it every day. They trust it.
Altus director Ben Hammond believes that tech including ChatGPT would not replace human advice as trust can only be built using transparent and engaging education and communication. And, Louis Williams, head of psychology & behavioural insights at Dynamic Planner, focuses on empathy.
AI may not be able to provide emotional support in the traditional way but it does understand this and is learning and developing constantly. In fact, tech can detect human emotions better than humans – and it’s been used for years in the insurance market. Further still, some may not want to show their emotions to their advisor but would be much happier to do so with an avatar.
Trust and emotional intelligence are not something you have or don’t have, but something you can learn and develop. In both of these areas, this is precisely where generative AI and ChatGPT can rapidly develop and deliver.
Well, Robo-advice never really took off
For those that think this of no concern and cite the demise of the current “robo-advice”, then it’s important to get up to speed and understand the difference.
What was (or is) “robo-advice”, and what’s coming shortly, surmounts to a comparison of Wall-E versus the Terminator.
Robo “advice” falls down because it’s not advice. And this is predominantly why it hasn’t engaged consumers with larger sums. In its current form it doesn’t satisfy the more complex analysis of all of a client’s holdings (investments, pensions, property, earnings etc) that are required to give advice.
So, assuming the new “system” can do this what else is there to hold you back?
If you were a prospective client seeking advice with £250,000 which option would you choose – A or B?
Option A – Wait upwards of 6 weeks to receive an idea of how much you will receive in retirement. Pay (on average) over £1543 (the cost of advice, Nextwealth) and then pay £2,500 per year for ongoing service; or
Option B – Get the answer within 10 minutes and pay sub £50 upfront with an annual fee of sub £200?
Human Adviser or Avatar Adviser?
Currently, the “interface” is a human adviser. But what’s to stop that from becoming an avatar adviser, that you can interact with?
Robo that can deliver advice in a holistic manner will deliver in a much greater way than the robo-“advice” that we have experienced thus far. So, if it’s a choice between paying £250 upfront & £250 ongoing versus £2000 upfront and £1000+ per year with a system that can advise me in the right way then it’s likely more people will choose the former.
Conclusion
The successful advice firms of the future are already incorporating generative AI into their proposition and see the future advances. They see it is there to support them and enhance their proposition for the clients of today and tomorrow.
The sooner people start exploring the technology’s merits, the sooner their true value will be understood. It is in advisers own interests to embrace, rather than fight the tide.
Recently we’ve seen Schroders head of UK intermediary solutions Gillian Hepburn discussing advisers needing to ensure they keep an open mind. Schroders Pulse Survey showed that more than a quarter (27%) of advisers will ‘never’” incorporate AI in their advice process. Those firms are literally setting themselves up to fail. Meanwhile, 43% of advisers see it as a potential threat. And they are right if they do nothing about it.
Advisers WILL be replaced by generative AI/ChatGPT but not those that embrace it. Roles will change as will the ratio of advisers to paraplanners. The dynamics of each firm type will change. Advisers will be able to undertake more client meetings, whilst paraplanners will be able to support many more advisers than they currently do.
Advisers will be able to do six meetings a day and thus double or triple their client base. Paraplanners will be able to support far more clients than they currently can, just like advisers. Some firms will see this as an opportunity to grow whilst others will see it as an opportunity to cut significant cost.
And for those who choose to ignore technology developments the new Advice-Robo platforms will attract a much wider number of clients with more significant assets than its baby brother Robo-“Advice”.
