SA's multi-billion telecom upgrades risk AI setback, report warns
According to Cloudera's Data Readiness Index 2026, many organisations are investing heavily in infrastructure while struggling with data accessibility, integration and governance, issues that could undermine AI-driven transformation initiatives.
The warning comes as local operators continue rolling out major network investments, including Vodacom's R85.2bn Vision 2030 programme and MTN's efforts to expand 5G coverage, modernise its network and increase the use of artificial intelligence (AI) across its operations.
The report argues that while these investments will provide the physical infrastructure needed to support AI applications, fragmented and poorly governed data could prevent operators from fully realising the benefits.
Data challenges persist
Cloudera's research found that 90% of organisations in Europe, the Middle East and Africa (EMEA) plan to increase cloud spending, compared to a global average of 65%.
However, despite 89% of EMEA IT leaders saying they know where their data resides, 42% reported that complex access processes remain a major obstacle to using it effectively.
Globally, 60% of telecommunications leaders said they could not access the precise data needed to make critical operational decisions.
The study also found that only 34% of EMEA organisations have fully integrated data across hybrid environments, leaving many businesses with fragmented datasets that create delays and increase cloud costs.
According to Cloudera, these shortcomings could limit the effectiveness of AI applications such as predictive network maintenance, fraud detection and customer experience optimisation.
Governance key to AI success
The report also highlights data governance as another significant challenge. Only 26% of EMEA IT leaders believe their enterprise data is fully governed.
For South African telecommunications companies, this presents additional risks because AI deployments must comply with the Protection of Personal Information Act (PoPIA).
Athul Prasad, global director of AI industry solutions for telecommunications, media and entertainment at Cloudera, said physical infrastructure alone would not determine AI success.
He said operators also needed secure, integrated access to their data across private and public cloud environments if they wanted to maximise returns on their AI investments.
Cloudera recommends that operators adopt a private AI approach, where AI models are deployed close to where data is stored rather than moving large datasets across public cloud environments. The company says this can improve data security, support regulatory compliance and reduce cloud-related costs.
The global study, conducted with Researchscape, surveyed 1,270 IT leaders across the US, EMEA and Asia-Pacific regions during early 2026, including 312 respondents from the EMEA region.
