There is an ongoing debate in all walks of life — seemingly — about artificial intelligence (AI).
Its uses, its pitfalls, and where it can save us all time, effort and, ultimately, money.
The discussions around this seem to have picked up some traction in the past year or so, particularly with software like the ChatGPT language model giving the everyman access to such technology where it may have otherwise been out of reach or, frankly, of not much use in your day-to-day.
In the specialist finance sector, and in bridging in particular, we have already seen advances in non-human based tech to enable quicker transactions, a cheaper route to finance and a slicker process for everyone involved.
Can common-sense lending and good rates not live happily ever after?
Uses such as online valuations, desktop valuations and the like are a great way for a client to be accepted by a lender or bank, and to move from application to offer in what feels like lightning speed.
My concern, however, is that, as we move forward, are we likely to rely too heavily on AI and to move away from ‘common-sense lending’, which every lender professes to offer?
Shackles
Tech/AI offers us a great opportunity to progress deals where it makes sense to do so. However, from speaking to colleagues and industry peers we can already see the shackles it imposes.
My concern is that, as we move forward, we may rely too heavily on AI
We have come across more than a handful of situations where we found ourselves on the end of phrases from lenders such as, ‘We would love to do this deal, but…,’ or, ‘We know the deal makes sense, but….’ Essentially, they are constrained by their own criteria.
Admittedly, in all cases that qualify for online valuations, for example, far more benefit from this than do not.
That said, it’s always a rather difficult conversation to have with a client when they are looking to borrow <10% loan-to-value and we need a full valuation (an expensive one, at that).
Conversations around why a lender is declining a deal, even though it knows it makes sense, are more common than ever
We should be striving to use automation of any kind to make our industry better wherever we can — but not at the cost of humans making common-sense decisions.
Many case studies and memorable deals come off the back of someone at the lender or bank taking a look at a full case, understanding the pros and cons, and finding a way to make a deal work.
I fear this may become a dying art if we rely too heavily on AI and tech.
From speaking to colleagues and industry peers, we can already see the shackles tech imposes
I appreciate, of course, that funding lines have certain cans and cannots. However, lenders that give humans a mandate of some kind to make decisions when it benefits them and the borrower seem to thrive on tricky cases.
‘It’s a bridging loan. Every case is tricky, isn’t it?’ I hear you cry. Unfortunately, conversations around why a lender is declining a deal even though it knows it makes sense, and in theory it could fund it but it doesn’t, are more common than ever. Just because it doesn’t tick the box.
You may think I am over the top here, and perhaps I am. However, there have been lenders set up in recent years off the back of seeing the art of common-sense lending slowly submit to the criteria-based ‘Computer says no,’ conversation, and they have come to the table to fill that void; funding deals that are a little outside the criteria but which make sense. Some lenders have no criteria at all.
We have already seen advances in non-human based tech to enable quicker transactions, a cheaper route to finance and a slicker process for everyone involved
That is great to see, but this flexibility ultimately comes at a cost, and the borrower foots the bill for that cost.
Maybe I’m a romantic, but can common-sense lending and good rates not live happily ever after?
Give the power back to the people.
Sam O’Neill is head of bridging at Clifton Private Finance
This article featured in the June 2023 edition of MS.
If you would like to subscribe to the monthly print or digital magazine, please click here.
