AI is one of the key words in the current online environment and has been for several years now; from the growth of LLMs and other generative AI to the spread of AI-powered tools, artificial intelligence has gradually spread to open new possibilities across multiple fields and markets. At the same time, AI adoption drives data increases – in many cases of single-use data, ballooning companies’ data usage and storage statistics.
Some of the most striking data on this question comes from storage and data management supplier NetApp, which questioned more than 1,000 data managers or architects in the UK in January 2025.
The survey found that businesses expect their data footprint to grow by 50% as a result of AI projects, although it also found that a significant portion of this data remains unused. More than one third (37.5%) of business data is believed to be unused generally, while one-fifth of respondents report that more than 50% of their data is unused.
This is a category known as single-use data, which has grown by a remarkable degree as a proportion of data generally thanks in part to adoption of AI. This problem is compounded by the – also AI related – perspective that datasets now have an intrinsic value in and of themselves, making many organizations reluctant to outright delete their information.
What does this data increase mean?
According to the survey, nearly one third (30.5%) find it difficult to identify what data to keep and what to dispose of because of the data’s potential use for AI projects in the future. Many also reported that the sheer volume of data they hold is a barrier to cleaning up data stores, with 30% saying it’s more economical to keep data than to try and clean it.
As such, AI adoption drives data increases while also presenting a barrier to rectifying the increase. As a consequence of this, companies will see data related costs – storage space on premise or the cloud, data management and data handling – skyrocket compared to previous levels.
Archiving is a known solution to this problem but runs into an already disclosed barrier. Across the survey, just under 27% said they lacked the resources to identify whether data is useful or not, while the same number cited a lack of budget or resources to manage data disposal. Without a way to identify data they no longer need to keep at all, or which could be archived to less costly media in the cloud or elsewhere, organizations are opting to ignore the problem as temporarily less costly to ignore than to fix.
In the long run, of course, this problem will only get worse as the data load increases. This is especially true if the current AI trend continues. Already it is commonly accepted that well structured and optimized data is a competitive advantage, both in cost structure, overhead, or risk mitigation.
The business need to optimize data
The wider issues associated with poor cost optimization are a challenge many businesses are just kicking down the road because they lack the visibility, strategy or tools to tackle it effectively. While this is understandable, it also reflects an unwillingness to confront the issues they are inevitably going to face.
One of the easiest ways to start is by addressing the tools aspect of the question: taking stock of the data management and storage systems your company already maintains, and what solutions are available to fill gaps you currently do not have covered.
An easy place to start is with TECH-ARROW’s Storage Optimizer, the Basic version of which is available for free. Using Storage Optimizer to sort your data gives you insights such as what file types make up the majority of your data load or what percentage has not been accessed in a given time period; this information can inform your data management strategy going forward regarding what information you need to keep on-site, what can be sent to an archive or cheaper storage on the Cloud, and what can be outright deleted.
No matter what steps companies want to take, it is critical that they begin planning it out now or at least as soon as possible. Like it or not, the AI revolution is here to stay; even if the newness of AI tools fades away and the rate of adoption slows, which it is likely to, AI tools are unlikely to fade away so long as they maintain some degree of utility. Companies need to get ahead of their AI-driven data increases now to stay ahead of the curve, or risk being swamped as the trend continues.
Your Data In Your Hands – With TECH-ARROW
