Should you trust AI to manage your finances - Y/N?

Chris Middleton Profile picture for user cmiddleton March 29, 2023
As the AI hype cycle spins ever faster and higher, should banks, or their customers, trust technology to help them be smarter with money?


Should you trust AI to manage your personal finances and investments? Tel Aviv-headquartered Personetics says you should. 

Backed by blue-chip investors such as Viola Ventures, Sequoia Capital, Lightspeed Venture Partners, and private equity giants Warburg Pinkus and Thoma Bravo – plus two banks not listed on the website, Santander and Asia’s UOB – the B2B AI company says it is here to make finance smarter. 

On the face of it, the formerly staid banking sector – which once worked on the tedious predictability of compound interest, but now seems populated with high-octane risk-takers – would seem to be the perfect platform for AI solutions. With the potential to help customers manage their income, investments, bills, and debts more safely and intelligently, backed by data.

But are AI companies like Personetics really working for those customers (135 million of them, it says) in a world of Open Banking and the rush to mobile fintech solutions? Or to help banks sell more product, at a time when trust in that sector is falling once again? 

Bank collapses and news of $300 billion in emergency borrowing from the US Federal Reserve certainly bring back painful memories of the last financial crisis. Is now really the right time for banks to sell, sell, sell to customers who are struggling with soaring inflation and flatlining wages? Perhaps it is, if AI and data analytics can help people to manage or invest what money they still have.

Dorel Blitz is Personetics’ VP of Strategy and Business Development. He admits:

We are on a mission to really help banks to efficiently sell to their customers and also, efficiently, to better engage with customers. The problem we are really trying to solve is helping banks to really get to know those customers. And by doing so, to engage and sell to them. 

What we have seen in the last 25 years, since digital banking really kicked in, is that banks don't really know their customers, and they definitely don't know how to engage with them. Or how to personalize every interaction.

This is ironic, as Know Your Customer (KYC) is a key regulatory requirement, albeit primarily from an anti-money-laundering (AML) and anti-financial-crime perspective. But Blitz certainly seems to know his customers: a good thing, from banks’ perspective. He continues:

On the flipside of that, I think that customers are really not just expecting, but demanding that their bank helps them to better manage their financial lives, to help them self-manage their finances. People are expecting their banks to become this real-time cool friend, a trusted advisor. 

But I think that banks don't really know how to live with customers’ expectations, to fulfil them. A lot of banks speak about personal engagement with their customers, but they are not doing it.

Fair enough. The reason, he says, is because it is hard for banks to “leverage their biggest asset” – their customers’ transactional data:

Think about it. You're making, I don't know, 20 to 25 different financial decisions a day. But the banks are just acting as this safe deposit vault and not managing your finances, actually managing all the data. 

But anytime you're making a transaction, anytime you're paying something or receiving money, all of this data is being stored within the bank, right? But they don't know how to use that, how to analyse it and help you in real time: your financial behavior, what are you’re doing with your money, your spending patterns.

OK. But is it really banks’ data to slice and dice, or their customers? While many people might appreciate AI-driven advice on demand, what evidence is there that customers want their banks to be more intrusive and – frankly – sales led? The Personetics website quotes 35-40% increases in customers’ digital engagement with their banks. Arguably, that’s more of a marketing focus than a personal money-management one. 

For UK bank customers specifically, that might bring back memories of, say, the hard-sell of unnecessary payment protection insurance (PPI). Not that that scandal can be blamed on Personetics, of course, but it certainly can be on banks’ desire to sell customers products they didn’t need.

Blitz argues that a more engaged and helpful bank is a more trusted one from the customer’s perspective, offering them a “financial wellness proposition”:

Banks are sitting on so much data, but they have lacked the ability to, in real time and on a personal level, analyse it, understand it, and engage with you. With different insights, allowances, or nudges, helping you to gain more confidence in the bank.

There is not really a one-to-one engagement in many cases, so customers feel that their banks don't know them, and are definitely not engaging with them. […] And that's all part of the ‘self-driving finance’ vision that Personetics has. It's very similar to self-driving cars.

Yes, but self-driving cars crash from time to time. As do banks. 

Whose data? 

Several times in our conversation, Blitz uses the phrase “our data”, apparently to mean Personetics’ own data. Is he suggesting that not only do customers’ data belong to the banks, but also to his company? He says: 

We are only looking at the same data, that you're saving today in your own mobile banking application, which is probably the last six months of historical financial transaction data.

Asked why then he described it as “our”, as in Personetics’, data, Blitz says: 

No, it's not our data, I apologize. We have no data. We have an engine, if you think about it that way. We are an engine, we are able – with a lot of best practices, capabilities, examples and use cases – to deal with any financial data, any transaction data. But we are using only the data already stored within the bank, which is your data.

But Personetics does look at my last six months of financial data, he adds:

Exactly, yes, because this is the data in the regulatory perspective, it's already being stored within the bank, and we have a white-label solution. Nothing is going on outside the page. We are analysing in real time the data that's already being stored within the bank.

The point here though is that I've trusted my data with my bank. I haven't trusted it to Personetics. Blitz counters: 

Yes, but you can say the same thing about Microsoft or IBM or Salesforce, or any other large IT company. Banks are using different vendors and technologies to serve you.

OK. But fundamentally, is Blitz certain that customers want a more intrusive relationship with their bank? Or do they simply want a low-friction service, one that doesn’t push them – via increased customer engagement – into what might be construed as a more marketing-led relationship? He says:

That’s a very good question, and I think it's chicken and egg. I think that, yes, customers definitely want, expect, and demand that banks definitely – with cost-of-living prices rising – are not just a basic commodity.

No one really likes to deal with FOMO – fear of missing out – all day long. But I think the customers really appreciate JOMO, the joy of missing out [on their own terms]. 

Trust, right? That's what banks are selling. They want a trusted advisor that will help them or guide them or advise them – or at least provide them with insights.”

Finance, meet AI

The conversation certainly comes at an interesting and dynamic time in the worlds of finance and technology: nervous markets, worried customers, and a popular hype wave surging towards AI adoption.

However, a recent survey suggests that banks themselves need to do much more to encourage responsible, ethical use of AI tools. A report from AI research company Evident this month says that “there is no evidence of responsible AI activity” in eight of the world’s largest banks: 

Research shows one-third of North America’s and Europe’s largest banks lack transparency and are failing to publicly report on the progress of their AI development. Banks across North America and Europe are failing to publicly report on their approaches to responsible AI development. 

SVB’s collapse, First Republic’s bailout, and escalating concerns about Credit Suisse have thrust the issue of banking sector transparency firmly into the spotlight once again, while also highlighting the need for many banks to overhaul and improve their approach to risk management.

However, Evident’s AI Index found that while AI is already used by banks for many critical processes, from authenticating customers to risk modelling, eight of the 23 largest banks in the US, Canada and Europe currently provide no public responsible AI principles.

Alexandra Mousavizadeh, Evident co-founder and CEO, argues: 

AI could be the key driver of better risk management and decision-making across the global banking sector. However, it is vital that banks develop AI in a way that meets high ethical standards and minimizes unforeseen consequences.  Our research found a worrying lack of transparency around how AI is already used – and how it may be used in the future – which could damage stakeholder trust and stifle progress.

My take

AI certainly offers a world of promising solutions to business customers. But the public should take steps to ensure that their bank also understands the ethical dimensions of these tools. And is protecting their personal data while providing services they actually want.

A grey colored placeholder image