Q&A: Where are you on your AI journey?
In this Q&A, Datacom Consulting Director Ash Valanidas discusses how businesses can assess their AI maturity and move from education and experimentation into operating, integrating and scaling AI.
Research from Datacom and ADAPT finds 80 percent of Australian CFOs say their generative AI investments are still being evaluated with limited returns, with none reporting results above expectations.
Seventy-seven percent of security leaders are confident in their visibility across risks and compliance, yet only 32 percent have a tested recovery plan.
Thirty-nine percent of organisations are exploring agentic AI use cases, but only 24 per cent have the data readiness to support AI initiatives at all.
Australian enterprises are investing heavily in artificial intelligence, scaling cloud strategies and beginning to explore agentic capabilities. Executive teams are under pressure to modernise core platforms, embed intelligence into operations and demonstrate returns, often simultaneously.
A new report from Datacom and ADAPT examines where Australian enterprises in high-stakes industries sit across modernisation, AI adoption, cloud strategy and operational resilience, and finds a consistent pattern: the ambition is there, but the foundations required to support it are lagging.
The report, AI ambition vs enterprise capability, draws on multiple ADAPT survey programmes, Datacom's 2026 Cybersecurity Index, and executive interviews across government, utilities, supply chain, logistics and manufacturing.
Among 104 Australian CFOs surveyed, 80 percent say their generative AI investments are still being evaluated with limited returns. Eight percent say investments have met expectations. None reported results above expectations.
Interest in agentic AI is building alongside generative AI investment, with 39 percent of organisations exploring use cases and scoping proofs of concept. But only 24 percent have the data readiness to support AI initiatives at all.
Mike Walls, Director of Cloud and AI Infrastructure at Datacom, says this pattern is something the team sees regularly.
"Ambition is rarely the constraint. Enterprise readiness is. We regularly see customers start modernisation not fully understanding the why and assuming they know their estate, only to discover undocumented dependencies, custom integrations, and operational realities that impact what's possible.”
Around 40 percent of mission-critical applications across Australian enterprises still depend on legacy platforms underpinning asset management, billing, logistics, finance and operational continuity. Modernisation progress across infrastructure and application measures averages between 30 and 44 percent.
These are mid-transition estates. Applications are layered, integrations are custom-built, dependencies between systems are often undocumented, and institutional knowledge sits with a shrinking number of specialists. Only 36 percent of the workforce has been upskilled in the past 12 months, and infrastructure leaders are anticipating a 24 percent increase in total power consumption over the next two years, pointing to a stressed operational environment.
Cloud strategy has moved into a more deliberate phase, with a quarter of organisations now actively reassessing workload placement across public cloud, private cloud and on-premises environments. But the state of cyber readiness reveals a striking disconnect.
Datacom's 2026 Cybersecurity Index found 77 percent of security leaders say they have sufficient visibility across risks and compliance. Yet only 32 percent have a tested recovery plan.
Mark Hile, Managing Director of Infrastructure Products at Datacom, says this distance between perception and operational reality is where some of the biggest risk sits.
"A plan isn't proven until it's been tested. As organisations operate across cloud, legacy, and increasingly AI-enabled environments, the cost of discovering weaknesses during an incident rather than beforehand can be significant.”
Datacom Director of Artificial Intelligence, Lou Compagnone, says the disconnect between pilot success and scaled deployment is one of the most common patterns across the market.
"I see this regularly – an organisation runs a successful AI pilot in a contained environment with clean data, then struggles to replicate those results across their broader estate where data quality is inconsistent and governance is still being defined.”
"The pilot worked because the conditions were controlled. Scaling is a different proposition. It needs the foundations in place first, and for most organisations, that work is still in progress. It comes back to going slow to go fast."
The report explores these dynamics in detail, including industry perspectives from energy, utilities, resources and regulated sectors, a maturity snapshot to help leaders assess where their organisation sits, and practical direction on where to focus before scaling further.
AI success depends on more than technology alone. Secure, scalable cloud platforms help organisations build the foundations needed to support innovation, manage risk and realise value from emerging technologies.
Learn how Datacom helps organisations get more from their cloud investments.
