IDC survey finds only 50% of service providers’ internal data is accessible to their AI models and analytics
65% of respondents to survey by IDC said AI has speeded up their move to cloud-first data architecture while 50% say they can only
Amdocs has released the findings of a global study it commissioned, conducted by International Data Corporation (IDC). The IDC report, AI Demands More: Enterprises Are Playing Catch-Up on Mission-Critical Data Modernization, highlights how “robust data modernization” is required to leverage the power of “hybrid” AI – that is GenAI and agentic AI.
IDC surveyed and interviewed of 151 execs within telecoms service providers across North America, Europe, the Middle East, Africa and Asia. The study will be published in full in March.
No good in isolation
The research found that respondents realise they cannot deploy GenAI in isolation and 78% of them said GenAI will need new investment in automated data cleaning and quality checks over the next two years.
“Effective AI solutions are built on trustworthy data, but our research shows that 50% of organizations still struggle to fully access the value of existing data resources and to ensure consistent data quality,” said Mary Johnston Turner, IDC Research Vice President for Digital Infrastructure Strategies.
“The emergence of powerful agentic AI technologies to orchestrate workflows and autonomous decision-making across multiple AI models will require most organizations to invest in data modernization to ensure the levels of end-to-end data health, security and accessibility needed to deliver reliable AI results.”
Key findings include:
- Some 65% of respondents have accelerated their shift to a cloud-first data architecture due to GenAI and increase investments in cloud computing, storage and GPU capacity to meet processing demands.
- As cloud modernisation shifts toward AI-centric architectures, service providers must balance scalability, security and cost control while strategically navigating data governance challenges and the high costs of AI infrastructure.
- Some 76% of service providers expect that service-as-a-software enabled by agentic AI will drive major or meaningful transformation across their organisation’s business model in coming years.
- Some 73% of respondents expect agentic AI to require more investment in modernising data as it reshapes business operations, enabling automated decision-making, dynamic forecasting and adaptive customer interactions across multiple departments. They range from sales and marketing to operations and cybersecurity.
- Unlike traditional AI models which function in silos, organisations are modernising their data infrastructure to support agentic AI-driven interoperability, using cloud-based platforms in pursuit of “seamless collaboration across use cases”.
- The “typical service provider” estimates that only 50% of internal data is accessible to its AI models and analytics.
“Enterprises increasingly recognize the importance of combining multiple AI capabilities – ranging from NLP to ML,  predictive AI to generative AI – in order to achieve scalable, accurate and valuable and dynamic business outcomes,” said Anthony Goonetilleke, Group President of Technology and Head of Strategy at Amdocs.
“However, reaching the full promise of GenAI means addressing data standardization, modernization and governance while embracing scalable cloud infrastructure. Companies that prioritize developing a solid data foundation today will be best positioned to drive GenAI’s long-term business impact.”