AI readiness illusion as data controls lag, Cloudera says

Cloudera has published research suggesting many organisations are adopting AI without the data controls needed to support it. The study surveyed nearly 1,300 IT leaders globally.

Its Data Readiness Index highlights a gap between confidence in data strategy and the ability to use and govern data across large organisations. In Asia Pacific, 82% of respondents said their data strategy was well designed and aligned with business goals, yet 38% said complex access requirements and processes were limiting effective data use.

The findings suggest enthusiasm for AI is outpacing progress on integration, governance and access. Across the region, 94% of respondents said they were very willing to adopt new governance frameworks to improve data readiness, but only 10% said all their data was fully governed.

Cloudera described this disconnect as an "AI readiness illusion", with businesses believing they are prepared to scale AI despite unresolved issues in how data is managed and shared. Globally, 96% of organisations have integrated AI into core business processes and 85% said they have a clear data strategy, but about four in five said their AI and data efforts were still constrained by limited access to data across environments.

In Asia Pacific, organisations appeared further ahead on some measures, but the data still showed weaknesses in integration. While 82% reported a clear data strategy, only 27% said their data sources were fully integrated.

Access problems

Data access emerged as a central obstacle. While 85% of Asia Pacific respondents said they had complete visibility into where their data resides, more than a third still struggled to use it because of access rules and internal processes.

That gap matters as companies try to move AI projects beyond trials and into day-to-day operations. AI initiatives most often fall short because of data quality issues, cost overruns and poor integration into existing workflows.

Among Asia Pacific respondents, data quality problems and weak workflow integration were the joint leading reasons for disappointing returns on AI spending, each cited by 19%. The results point to operational problems rather than a lack of interest in the technology.

Infrastructure also remained a hurdle. Globally, 73% of respondents said performance constraints had hindered operational initiatives. In Asia Pacific, 28% said infrastructure performance issues often hindered such work, while 38% said they sometimes did.

Governance gap

The survey also found a sharp divide between perceived data quality and formal control. Globally, 84% of respondents said they were confident in the accuracy, completeness and alignment of their organisation's data, but only 18% said all of it was fully governed.

In Asia Pacific, the governance gap was wider. Although many respondents expressed confidence in their data, only 10% said it was fully governed across the organisation.

This can leave businesses exposed when data is used across teams, systems and AI applications, especially where standards are inconsistent or data remains siloed.

"Ambition is high, but for many Australian organisations, AI is still running on a fragmented data legacy. In an era of sovereign data mandates and critical infrastructure protection, you cannot scale what you cannot govern. The path forward is turning trusted, well-controlled data into an operational capability before scale becomes risky," said Vini Cardoso, Chief Technology Officer, Cloudera Australia and New Zealand.

The regional findings come as governments and large organisations in Australia and neighbouring markets focus more closely on the use of AI in regulated and critical sectors. The survey suggests senior executives are aware of the issue: 92% of Asia Pacific respondents said senior leadership understands and prioritises the data infrastructure needed to enable AI at scale.

Responsibility for fixing the problem appears to rest mainly with technology leaders. In the region, 69% said chief information officers and chief technology officers are chiefly accountable for delivering data readiness for AI.

Industry split

The report also showed significant differences by sector. Telecommunications respondents reported stronger visibility and access to data than those in financial services and the public sector. More than half of telecommunications respondents said it was extremely true that they had full visibility into where their data resides, compared with about a third or less in financial services and the public sector.

On access, 51% of telecommunications respondents said they could access all their data at any time, compared with 24% in financial services and 16% in the public sector. Even so, telecommunications was also the sector most likely to say infrastructure performance consistently hindered operational initiatives.

Barriers to returns from AI varied by industry. Cost overruns were cited most often in energy and utilities, while poor integration into workflows featured more prominently in healthcare, manufacturing and financial services.

Cloudera's figures indicate that many large organisations have moved quickly to embed AI into business processes while still grappling with longstanding data management problems. The survey covered IT leaders at companies with more than 1,000 employees across the Americas, Europe, the Middle East, Africa and Asia Pacific.

"Enterprises aren't struggling to adopt AI, they're struggling to operationalise it beyond experiments. AI is only as effective as the data that fuels it. Without seamless access to all their data, organisations limit the accuracy, trust, and business value that AI can deliver. You can't do AI without data," said Sergio Gago, Chief Technology Officer, Cloudera.