iscover
Share this article on:
Powered by
MOMENTUM
MEDIA
The security teams of major Aussie bank Westpac are currently exploring the idea of using ChatGPT in their planning processes.
Richard Johnson, the chief information security officer (CISO) for the big four bank has said that artificial intelligence (AI) is a tool that needs to be secured by security teams from now on.
“Like every technology that’s come before it, [ChatGPT is] an advancement and something that we need to embrace, understand and work with,” he said at a roundtable at the Amazon Web Services (AWS) Summit Sydney.
“We need to understand how to secure that capability on an ongoing basis, and we need to look at the opportunities of how we can leverage that.”
Westpac is currently trialling use cases of generative AI and has identified opportunities for it in the banking sector for both security and finance.
“Within cyber security, we’re always focused on understanding how technology like AI can be used to improve cyber defences while also ensuring any use of it across our business is robust and secure with the right controls in place to mitigate attack risks,” he added.
Westpac is currently monitoring the risks of using ChatGPT and generative AI and says that it is “ensuring [it remains] well-informed and [is] building appropriate security settings early”.
The big four bank has shown interest in generative AI previously, after the company’s chief technology officer David Walker said last month that the technology had potential in the banking industry if it could “find its niche”.
At this stage, however, he has said it is “overhyped” and that with “no human-level understanding of concepts”, the AI can sometimes write reasonable-sounding but incorrect answers.
“So, is ChatGPT overhyped? The short answer is yes, for now, but generative AI is likely to be transformative,” Walker said.
“As the fanfare that surrounds ChatGPT begins to die down, this class of AI should find its niche in certain job functions and industries.
“For banks, generative AI offers many possibilities, from writing software and drafting policies, to using language to interface with applications and data.”
In an article published by AFR, Walker said that it would keep AI away from customers, as it has the potential to hallucinate and generate fake quotes, meaning it can never be fully trusted.
However, the bank has said that it is developing customer assistance chatbots using Amazon’s Lex systems.
Westpac first began exploring the potential of generative AI as it increased its ties with AWS, which would be responsible for providing the bank with generative AI technology it would use to write better emails to customers, while assisting software engineers with coding.
The two companies have signed a five-year deal to support its hybrid, multi-cloud strategy. It entered a similar deal with Microsoft last year to use its Azure cloud services.
Be the first to hear the latest developments in the cyber industry.
