Innovation and Growth Innovator Maile Carnegie on leading enterprise transformation through volatility

In this discussion at CIO Edge, Maile Carnegie draws from her experience across Procter & Gamble, Google and ANZ to share her perspective on Australia’s readiness for AI adoption with Alan Thorogood, Research and Engagement at MIT CISR.

While Maile sees strong potential in smaller private companies, her confidence drops when it comes to large publicly listed organisations.

She argues that Australia’s corporate environment is structurally geared towards stability rather than disruption.

This makes it harder for boards and executives to embrace the courage, risk appetite, capital access and talent needed to pursue agentic AI and large‑scale transformation.

By contrast, Maile notes that US public companies operate in ecosystems designed for growth, experimentation and rapid change.

There are several systemic factors that limit risk tolerance in Australian organisations.

Board directors here face significantly higher personal accountability, including exposure to stepping‑stone liability, and are responsible for a broader set of “social licence” expectations compared with their US counterparts, who are primarily held to financial performance.

Australia’s principles‑based regulatory environment also creates uncertainty: while experimentation is easier to start, organisations have less confidence that they won’t encounter regulatory or legal challenges later.

Access to growth capital is another barrier.

This is driven by franking credits that push shareholders towards yield, and by Australia’s low‑cost, highly accessible shareholder activism environment, where influencing corporate decisions requires “100 votes and $50”, compared with billions in the US.

Given these constraints, Maile’s advice for Australian technology leaders centres on three priorities.

First, she stresses the need to “chunk” AI initiatives into one‑year horizons, enabling clear and defensible returns rather than multi‑year promises.

Second, she urges leaders to focus on building trust by helping boards and executives manage compliance risks, particularly in explainability, observability and governance.

Finally, she emphasises the importance of deep partnership with business unit leaders: agentic AI demands not only workflow knowledge but also the “judgemental decisions” and contextual understanding that sit within business teams.

To succeed, organisations must align storytelling, strategy and operating models.

This requires shifting from human‑centred architectures to capability‑centred ones, and supporting business leaders through curated information, shared language and a unified vision for where AI‑enabled business models are heading.

Key takeaways:

• Australia’s large public companies face structural barriers to AI adoption, with boards operating in an environment geared for stability rather than disruption, reducing their risk appetite compared with US counterparts.
• Board accountability and regulatory settings limit courage and experimentation, as directors face higher personal liability, broader social‑licence expectations and uncertainty created by principles‑based regulation.
• Tech leaders must adapt by proving value quickly, chunking AI programs into one‑year ROI horizons, strengthening governance and compliance (observability, explainability) and building deep, trust‑based partnerships with business leaders to unlock judgement‑based workflows and capability‑centred operating models.