Bain & Co Head of AI Practice, APAC on how Aussie organisations can close the AI value gap

Bain & Co’s Richard Fleming talked about why Australia lags in AI maturity and how to move from pilots to real business value in this Data & AI Edge interview.

He highlighted the relatively slow progress of AI adoption in Australia compared to the United States and parts of Asia.

While many Australian organisations remained stuck in pilot phases and uncertain about return on investment, US firms reported that most of their AI projects were meeting or exceeding expectations.

Richard described a common issue: widespread but uncoordinated AI experimentation, often referred to as shadow AI.

The challenge, he said, was moving beyond isolated pilots to scaled implementations that delivered measurable business value.

Achieving this would require stronger governance and structured capability-building across the organisation.

He stressed the need to integrate AI efforts with existing business and technology functions, rather than treating them as separate initiatives.

Importantly, he encouraged leaders to begin real-world AI application development using the data they already had, in parallel with efforts to improve data quality and infrastructure.

Richard also noted that organisations often undervalued their unstructured data, such as customer conversations and internal documentation.

He advocated for shifting away from broad, slow-moving data programs in favour of use case-led strategies, an approach also reflected in ADAPT’s research, which found that productivity gains from AI were highest when supported by innovation culture and strong user adoption.

To embed AI effectively, Richard suggested creating dedicated leadership roles, similar to how Chief Digital Officers emerged a decade ago.

He also called for updates to enterprise risk frameworks, especially in regulated sectors, to manage emerging concerns such as model accuracy, compliance, and governance.

Ultimately, he emphasised that bridging the AI gap in Australia would require consistent dialogue between business leaders, boards, and regulators to align innovation with risk and responsibility.

Key takeaways

• Australia lags in AI maturity – Many organisations remain in pilot purgatory, struggling to scale AI and realise consistent business value.
• Use existing data to start building – AI success does not require perfect data. Real progress comes from applying AI while maturing your data capability in parallel.
• Create dedicated AI leadership and guardrails – Organisations need AI-focused leadership, updated risk management, and collaboration between business and technology teams to scale responsibly.