Three things every organisation adopting AI needs to be mindful of today

As Australian organisations embrace AI-driven change, three factors will be important to optimise their forward-facing strategies

If there's one thing that the last couple of years has taught us, it's that AI isn't about quick wins. 

These may have been viewed as important in the early days when AI's value in the enterprise wasn't as universally recognised. But AI adoption has now reached near-ubiquitous levels: in one recent study, every company was found to be adopting AI in some capacity. 

Today, rather than focusing on quick wins, Australian organisations understand it's much more about laying a foundation for measurable, sustainable and scalable change, and making sure that this change is implemented in a responsible way.

To achieve this, what's needed is a responsible approach to AI adoption, where responsibility covers the full spectrum of environment, social and governance (ESG) - from having a sustainable approach to AI adoption, to adopting key principles for social license to operate, and having strong foundations and platforms in place for data governance and overall enablement.

A principled approach to AI usage

Among organisations there's broad consensus on the need to have a way to build and run AI in a principled way: to make sure that every AI use case that gets stood up maintains that principled approach from design to implementation to operation.

Principles to some extent are defined already in large listed organisations and governments, although the rapid pace of change in AI means that these principles often require frequent revisions, and new principles may be required in order to embrace new AI use cases. 

At a country level and international level, there are attempts to unify countries under open, ethical, and inclusive AI principles. Australia is a signatory - joining over 90 other countries - and also has its own guidance at a federal level. 

For signatory nations to the international principles, the important thing to consider is how these principles can be implemented back home. These principles tend to be non-binding at present, but form a key part of an organisation's social license to operate. 

How long before all AI adopters will need to adjust their internal principles to reflect international standards, and will the standards be an upgrade on what companies are already doing in the space?

Organisations can't control what happens at a policy level, however they do have control of their own domains. What they can control is their own principles that meet users', customers' or stakeholders' expectations, and having other appropriate foundations in place for AI-enabled change.

Focusing on foundational data infrastructure

There is plenty of mileage left in data and platform-based improvements to enable AI-based change in Australian organisations. 

AI is only as effective - and as safe - as the data and platforms behind it, and while a lot of work has been undertaken on the foundational elements of AI in the last couple of years, more work is clearly required.

A recent Hitachi Vantara report revealed that 41% of IT leaders in Australia believe data quality is the most important factor in AI success, yet many still operate on fragmented and siloed data. This isn't just a technical bottleneck, it's a trust issue. Without clean, reliable data, AI decisions become opaque, error-prone, and hard to audit.

The use of hybrid cloud platforms, industry specific AI use cases and digital services offer a potential blueprint for success. Hybrid cloud infrastructure, a combination of on-premises and public cloud, is emerging as a clear favourite among organisations for building and deploying AI solutions. This preference extends to data pipelines used for moving and managing data. 

Additionally, embedding industry-specific capabilities into an AI solution stack ensures more accurate and relevant business outcomes, empowering organisations to harness the power of AI, regardless of where their data is located.

By building a strong foundation that includes a secure and scalable data infrastructure, organisations can unlock the true power of AI.

Making AI-driven change sustainable

Sustainability is increasingly a part of a responsible AI strategy. It has been a big topic this year as more efficient AI large language models (LLMs), in particular, emerge. 

Improving software efficiency is one aspect of sustainability; others include the use of AI-driven cooling systems that reduce power usage in data centres, and using efficient IT infrastructure to power and underpin AI workloads.

Given the rapidly evolving nature of the AI landscape, sustainability is not the highest priority for many organisations, but there are signs that it is increasingly on the strategic radar. A survey by Hitachi Vantara last year found 28% of Australian organisations rank sustainability as a priority in AI implementation. 

However, there is a demand for environmental sustainability and energy-efficient storage solutions. An equal proportion of IT leaders said they needed assistance to create scalable, future-proof storage hardware solutions that are secure, highly available, and efficient to meet sustainability goals. These form part of the hybrid cloud platforms that underpin AI efforts. Effective data solutions, it was noted, bring data closer to users while emphasising security and sustainability. 

In order to achieve more sustainability, one opportunity could be the use of shared or standardised models that enterprises could collaborate on, so that they can avoid reinventing the wheel and compare the efficiency of their AI architecture and power consumption.  This could involve reducing work between models or having a standard common model funded by a consortium of companies, government agencies or industry regulators ultimately enabling enterprises to understand what good likes like and what can I do better?

Conclusion

For organisations navigating the complexity of the AI landscape, the smartest path forward mirrors Hitachi's strategy: Think big, start modular.

Build trust through transparency. Prioritise sustainability as a business enabler, not an afterthought. And above all, ground AI in real-world use cases that solve real-world problems. This will give Australian organisations the best chance of being able to use AI to achieve measurable, sustainable and scalable change.