Siloed data hindering organisations’ AI projects

More than four in five (86%) organisations have experienced AI project delays or cancellations due to issues with their data, research commissioned by Iron Mountain indicates.

A survey conducted for the company by Vanson Bourne found that while 95% of organisations nominate using data to improve business outcomes as a top priority, 81% believe they will not be able to scale AI initiatives past pilot projects without first transforming their siloed and unstructured information into AI-ready data.

The research found that four in 10 organisations are spending at least 41% of their AI project timelines on manual data preparation. By contrast, only 39% are categorised as digitised and structured. These are organisations where dark data is centrally managed, indexed and readily usable for AI and analytics.

Meanwhile, 56% of respondents cited information spread across disconnected or siloed systems as a major barrier to AI projects, while 53% cited legacy formats or physical archives as a significant barrier.

In addition, 70% of organisations say they cannot effectively track the origin or journey of the data they use in their AI models, which threatens to be a big issue as businesses move beyond generative AI to agentic systems capable of acting across workflows, according to Iron Mountain SVP of Global Product and Solutions Marketing Varun Gupta.

“AI readiness is about much more than deploying the latest technology. Enterprises need to know that the information powering their AI is accessible, current and governed, with clear ownership and permissions,” he said. “As AI agents move from generating answers to taking actions across business workflows, these foundations become essential to scaling AI with confidence.”

Image credit: iStock.com/Vertigo3d