Home
You’re currently viewing our content for All Regions.
You can switch to a different region:
July 7, 2025
“There’s a bank that’s committed $100 million to AI over three years. There’s another organisation investing $24 million to transform insurance processes.”
Lisa Bouari arrived at the point quickly.
“Yes, there’s some real money behind AI in Australia now.”
As Regional AI Leader of Oceania at EY, Bouari’s commentary is delivered through a lens of objectivity and credibility – this is an influential leader who has been there, done that and got the t-shirt on AI.
“I’ve been waiting for 20 years for AI to become ‘a thing’ and now I’m certainly in the thick of it,” Bouari shared.
Drawing on a career spanning over two decades, Bouari is a former AI and data architect who started at Oracle before moving to IBM as part of the IBM Watson team. This is alongside co-founding OutThought in 2017, a start-up that specialised in developing conversational AI and digital human technologies.
“I’ve watched the hype of AI evolve over a long time and there is something dramatically different about today, in this era,” Bouari observed. “The use cases are real and organisations are putting real budgets towards AI.”
Bouari shared the stage with Devesh Maheshwari – CTO of Lendi Group – to headline the session AI in Action: Use Cases & Examples at Moxie Authority 2025.
Aligned to the theme of Inspired Knowledge, Moxie Authority 2025 housed the most influential figures setting the market agenda in business and technology across Australia.
This inaugural and invite-only conference in Sydney hosted more than 400 industry front-runners spanning all ends of the ecosystem, from CIOs, CTOs and CISOs to CDOs, CEOs and Founders.
Understanding AI use cases
According to Moxie Research, 69% of Australian businesses will identify “high impact” AI use cases in 2025, while 64% will deploy small pilot AI projects.
At present, the top five AI use cases in play across organisations in Australia can be ranked as:
Findings from Moxie Research mirror on-the-ground sentiment shared by Bouari when advising businesses on the best path forward to maximising AI.
“Use cases are very specific and dependent on industry but we are seeing patterns emerge,” Bouari explained.
“Through the advent of GenAI, we’re seeing lots of AI implementations around knowledge retrieval – policies, product information etc. Think of all the unstructured documents that exist within an organisation – contracts, contract comparisons, sales to invoice orders, payments, procurement, automated customer applications etc.”
Rather than adopting AI for novelty purposes, strategic organisations are prioritising outcomes – knowledge access, efficiency gains, revenue growth, risk reduction and experience improvements.
Based on Moxie Research, 41% of Australian businesses currently have a “systematic process” for measuring AI performance. That percentage is expected to increase as more use cases emerge and return on investment (ROI) requirements tighten.
“Take pharmaceuticals, if you can run a clinical trial three years faster through AI, that equals billions of dollars at the end,” Bouari added. “Or mining, a lot is happening in predictive analytics – such as budgeting and forecasting from the finance office.
“There’s also a huge uptick in coding and migration. We do a lot of work in on-premises to cloud migrations and now use AI in the coding which is resulting in x8 faster project completion.”
On the one hand, the use cases are pinpoint and specific. On the other hand, they are far and wide.
So, take a step back and let the problem shape the solution.
“We started by solving one of our most critical problems,” Maheshwari shared.
“If you look at a typical 9-to-5 day, how feasible is it that people pick up calls during those hours? You can tell by the number of missed calls you have yourself that it’s not always possible to respond.
“So we put the business problem at the centre and started solving this for our customers through an out-of-hours AI voice agent designed to free up time for brokers.”
Under the banner of Voxi, the product records and automatically transcribes calls before creating summary notes for brokers with action points and suggestions incorporated.
This then morphed into Voxi Coach.
“When we analysed our associates, they were getting no coaching,” Maheshwari expanded. “We have a lot of associates who are progressing from graduate level and want to become future brokers – and they want to learn.
“But no coaching existed so through Voxi Coach we now provide real-time call coaching. When a call is happening with the customer, a transcription is happening at the same time to give brokers a score and offer prompts. This has proved invaluable.”
The example shared by Maheshwari strikes at the heart of the AI debate – business problem first, technology solution second.
“Every field is becoming advanced because of AI usage but how do you use it in the right way?” Maheshwari questioned.
“In our case, how do we put it in the hands of our brokers in an effective way? We have 30+ lenders on our panel, each with different policies and types of products – some are complex and some are simple.
“That’s why we also created an internal chatbot for brokers to ask us questions about different policies while they are conversing with a customer. They receive an answer which means one less appointment and better customer service.”
Challenges… Mindset? Money? Mismatch?
As one leading CIO at Moxie Authority 2025 put it… “I’m not going to spend $5 million solving a $5 problem.”
And nor should any business. But as Bouari challenged, that’s not the best way to position the value of AI.
“There’s a legacy problem happening in market,” she continued.
Recalling time as a data scientist, Bouari referenced the historic tendency of organisations to take a “massive amount of data” and start blindly building machine learning models in an attempt to extract value from the information.
“It’s the equivalent of digging for gold without ever doing the homework in terms of surveillance and outlining basic principles such as where best to dig,” Bouari added. “That’s happened for decades with AI and I’m still seeing it today.”
For example, some large institutions in financial services are running 1000 use cases which is nothing more than “technology looking for a problem”.
“That’s a sure-fire way of ending up with a proliferation of POCs which inevitably ends up as a POC graveyard,” Bouari stated. “It becomes a whole set of science experiments that go nowhere and are very expensive – such an easy way to burn money in the business.”
According to Moxie Research – and by order of severity – the top AI project deployment challenges currently facing organisations across Australia are:
“The businesses putting their money where their mouth is on AI are the ones that have carried out due diligence on where the value is,” Bouari said.
As outlined by Bouari, this includes probing questions such as:
“In these instances, it becomes evident where to invest to achieve a return with AI,” Bouari advised.
“Because the alternative is an abundance of POCs that don’t end up going ahead because technology teams can’t demonstrate value and therefore fail to secure the funding. The board will always say no.”
Based on Moxie Research, CIOs are committed to the following AI initiatives during the next 6-12 months to:
Within this context, the key pillars underpinning AI strategies in Australia are:
“One of the challenges is demonstrating how good your data actually is,” Maheshwari cautioned. “You’re building on top of your data assets so if they’re not very good to begin with, then you won’t achieve an outcome and that leads to POC failure.”
In reference to Voxi, Maheshwari said the product originated from an internal hackathon meaning the POC was created within a day.
“It’s not expensive,” Maheshwari added. “Plus, it allows your teams to learn and grow all while keeping the problem that we were trying to solve at the centre.
“Many organisations are running on the premise of ‘let’s adopt AI for cost cutting’ but in my opinion, that’s absolutely the wrong place to start. AI is a technology enabler to reach an outcome, not the outcome itself.”
Secondly, Maheshwari cited the importance of technology modernisation as a starting point for any future AI deployments. According to Moxie Research, 40% of CIOs most closely self-identify as a “moderniser” in today’s market.
“If you do not modernise or disrupt yourself, then somebody else will do that for you,” Maheshwari warned. “Yes, it’s a cliched line but that’s how we built Voxi.”
In sharing the step-by-step process of product creation, Maheshwari highlighted the high conviction levels required to pursue innovation at pace.
One of the core values running through the corridors at Lendi Group is a “hell yes” mentality. No project or program of work is started without ticking the “hell yes” box first.
“That means we have full conviction and believe we will succeed,” Maheshwari detailed. “We do all of the groundwork before that with our product teams, customers and brokers to ensure that the project will work.”
The first build of Voxi cost the business approximately $100,000 with roughly $1200 per day being burned during the process. The second iteration however was constructed within a week and now costs $50 per day.
“That’s the price for learning but when you start to build, you have to make a choice,” Maheshwari said. “So many prototyping tools now exist that allow your teams to play with products in a low cost way.
“Take Lovable.dev or Magic Patterns as examples. I’m not endorsing those products but you can test out the value for a small license fee of about $20. Test them with your designers and your customers. That’s the cheapest form of learning.”
On AI, heed the expert advice
As AI moves from hype to practical application, businesses are increasingly focused on identifying high-impact use cases that deliver measurable value.
This has triggered a strategic shift. AI initiatives are now being evaluated through a commercial lens, with executive leaders exploring where and how this technology can make the biggest difference.
“What’s the one narrative in market that’s wrong about AI?” Bouari asked.
“That it’s high risk. People equate not understanding the mathematics or deep dark depths of an AI model as being risky.
“If you have a good team that implements the guardrails and asks the right questions, then you deploy it in smart and safe use cases, then it’s not high risk.”
On the contrary, Bouari stressed that “it’s actually riskier not to embrace AI”.
With EY as a case in point, the consulting giant has committed to training 400,000 employees worldwide to help stay ahead of new market disruptions on the horizon.
“It’s not difficult,” Bouari said. “Provided you have the right pillars in place and take pragmatic and practical steps rather than being too ambitious from the outset.”
For many businesses, the race is no longer about being first, but being focused – finding the intersections where AI aligns with strategic goals, delivers tangible benefits and builds a foundation for sustainable transformation.
“Identify your use case and get your hands dirty,” Maheshwari advised.
“This is also for the leaders and not just the teams. My biggest learnings came from TikTok videos, which I then implemented. Which is the second part – materialise that learning.
“There is an abundance of free content available today so find time to up-skill yourself and your teams within a non-production environment with privacy guardrails on customer data.”
In other words, as Maheshwari closed out – “let them play”.
Moxie Authority 2025 housed the most influential figures redefining business and technology across Australia. This inaugural and invite-only conference in Sydney hosted more than 400 industry front-runners spanning all ends of the ecosystem, from CEOs, CIOs and CTOs to CDOs, CISOs and Founders.
Inform your opinion with executive guidance, in-depth analysis and business commentary.
