OpenNashPodcasts

Person

Hani Hagras

Hani Hagras is a speaker from Temenos who appeared on a podcast about AI and the future of compliance in financial services. He talked about why banks need AI they can explain, and where it can be used in areas such as credit decisions, fraud checks and customer retention.

Hagras said customers turned down for a loan have the right to understand why. He said his team has worked on explainable AI for about 20 years and holds several patents. He argued that black-box systems are fine for tasks like speech recognition, but not for banking decisions.

1 episodes · 13 quoted claims

Hani Hagras says cheap data storage and efficient databases are among the developments making AI necessary in banking.

“Now it is very, very cheap to store huge amount of data. and also we have got very efficient databases which can easily organize such amount of data.”

AI in Financial Services · 0:02:46

Hani Hagras says black-box AI is acceptable for tasks like speech recognition, as with an Amazon Echo Dot, because nobody needs to know why it works.

“as long as your Amazon dot is doing very well as to colonizing your speech, you don't really care what is happening in the background”

AI in Financial Services · 0:05:40

Hani Hagras says customers rejected for a loan or financial product have the right to understand why.

“if you have been rejected, for example, a certain loan or a certain financial product, they have got the right to understand why I have been rejected for this.”

AI in Financial Services · 0:06:25

Hani Hagras says the UK House of Lords select committee on artificial intelligence stated that AI must be intelligible and explainable to be trusted.

“They put it in black and white that if artificial intelligence is going to be accepted as a trusted entity in our society, it needs to be intelligible, needs to be explainable.”

AI in Financial Services · 0:06:55

Hani Hagras says his team has worked on explainable AI for about 20 years and has been granted several patents for a method of generating rules from data.

“We have been working in this area for about 20 years. And one breakthrough that we managed to do, and we have been granted several patents out of this”

AI in Financial Services · 0:09:13

Hani Hagras criticizes model induction systems because a black-box model still makes the decision while a second model tries to interpret it.

“what's called model induction systems which basically still rely on a black box model running in the background and try to interpret it by another model”

AI in Financial Services · 0:10:28

Hani Hagras says Temenos's explainable AI platform can take 600 to 1,000 features about an applicant from a credit referencing agency for credit decisions.

“we are able actually to take 600 or 700 or 1 ,000 features about yourself, which come from any kind of credit referencing agency.”

AI in Financial Services · 0:12:02

Hani Hagras says Temenos's platform produces models with few rules, each only three or four conditions long, even when the model draws on about 1,000 inputs.

“generate from this kind of data models that are having few rules where the length of each rule is only three or four antecedents although that your models might be using 1000 inputs”

AI in Financial Services · 0:12:31

Hani Hagras describes a cash flow prediction model that looks at a customer's transactions and predicts a deficit over the next six months.

“If there is a model looking at your transaction and is able to predict that in the future, in the next six months, you'll be in deficit.”

AI in Financial Services · 0:13:31

Hani Hagras argues that predicting customer churn with a neural network is of little use unless the bank also knows why customers leave and how to keep them.

“if you speak about something like churn, there is no point of me just predicting that this kind of customer is going to churn by a neural network and that's it.”

AI in Financial Services · 0:15:13

Hani Hagras says knowing why a transaction was flagged as fraudulent helps banks reduce false positive rates.

“And what is important as well to understand why this kind of prediction has been flagged as fraudulent so that basically we can reduce the false positive rates.”

AI in Financial Services · 0:16:00

Hani Hagras names credit decisioning as the easiest early use of explainable AI in banking.

“the low-hanging fruit for this would be definitely credit decisioning, because everybody would like to understand why this kind of score is like this and what it would take them to improve it.”

AI in Financial Services · 0:18:29

Hani Hagras says customers cannot spend hours at a bank branch for tasks that can be done on a mobile phone, which drives interest in conversational banking.

“People cannot afford actually to go to the bank for two or three hours and lose so many times of the day to see somebody on something that can be done on their mobile phone.”

AI in Financial Services · 0:18:58