Aussie AI Initiatives Blocked By Tech Debt: What To Do About It

Last month, the Australian government launched a National AI Plan with the aim to help the country build a world-class AI industry that will capture the economic opportunities from AI. While there is massive potential for Australian businesses, recent IDC research revealed that not everyone is on equal footing.

In fact, there is an Aussie AI readiness gap.

According to the research, which was commissioned by MongoDB, 58 per cent Australian organisations are still operating on data architectures so inflexible and slow that building new applications without extensive modernisation is nearly impossible.

While some companies are rebuilding for the AI era with data architectures suited for the dynamic, unstructured data required to power AI applications, many organisations are trying to “bolt on” AI to outdated systems. The strategic choice made by the former leads to long-term AI success and growing digital revenue, while the “lift and shift” approach from the latter leads to rigid systems that cannot meet AI’s requirements.

Lendi Group is a great example of a company which made the strategic choice to build on a modern data infrastructure foundation. By fully embracing modernisation, they were able to move beyond AI experimentation and become a truly AI-native business.

Don’t Let Legacy Jeopardise Your AI Success

A high volume of quality and timely data is the real power behind successful AI applications. Most organisations have access to the same LLM models, so it’s the data that is the differentiator.

In many cases, legacy relational databases are a root problem – they are too rigid, costly and slow to meet today’s AI requirements. Legacy slows everything down, it adds complexity and siloes data, which in turn blocks the AI use cases.

IDC’s report actually shows that businesses across APAC who fail to address their legacy constraints risk a 50 per cent higher failure rate for their AI initiatives.

Not only that, modernisation is directly correlated to revenue down the line. Australian organisations that are prioritising modernisation generate nearly three times more digital revenue (68 per cent) than their counterparts who do not (24 per cent) – a testament to the transformative power of modern, AI-ready data systems.

Aussie healthtech scale-up Heidi is another great example here. Their strategic choice to build on a modern data foundation helped the company scale its AI Scribe to 81 million clinical consultations in just 18 months.

Four Essential Steps For A Successful Modernisation Strategy

Modernisation should be considered as an ongoing discipline rather than a one-off project. Time and time again, we’ve seen lift and shift approaches fail in the long term: AI and the data feeding it are constantly evolving, they need an infrastructure that adapts.

To achieve modernisation success, and unlock genuine AI innovation, organisations must focus on four main priorities:

1. Prioritise talent and skills development: Successful modernisation hinges on building and nurturing talent. Ongoing modernisation must support both human and technical resources within IT. AI skills are in short-supply, but training is becoming abundant, so ensure there are suitable programs that support the deployment plans.
   
   
2. Adopt cloud-native strategies: Future success depends on the flexibility to move workloads to the optimal platform, whether on-premises or in the cloud, as business needs evolve. For advanced workloads such as AI, cloud-native approaches are essential. Avoid repeating the mistakes of “lift and shift” migrations, which can create new technical debt and drain resources.
   
   
3. Accelerate data maturity: Data maturity is critical to avoid future setbacks. Accurate, well-managed data enables IT to align projects with business goals, meet compliance requirements and drive rapid innovation. Invest in platforms that support hybrid cloud environments without adding unnecessary complexity or risk. Furthermore, data modernisation initiatives should be embedded as operational procedures, instead of a one-time project.
   
   
4. Modernise database technology: Maintenance of legacy databases has proven to be a challenge to modernisation. Modern applications require modern database foundations. Upgrading database technology is key to improving data management and supporting the demands of today’s digital business. Ensure database solutions are scalable and flexible to accommodate future growth, new data types and evolving AI workloads. Opting for a flexible document model instead of a traditional relational model is a good start.

Many companies are still trying to make yesterday’s architecture carry tomorrow’s AI workload. But companies pulling ahead are the ones who have modernised and rebuilt the systems powering AI, and are now able to move beyond prototypes and into successful, scalable production and delivery.

Simon Eid is SVP, APAC at MongoDB

Last Updated on April 20, 2026 by Simon Eid