Report: Data access hurdles affect AI adoption for 71% of enterprises

Even as decision-makers and CXOs remain bullish on AI’s potential, enterprises are struggling to make the most of it at the ground level. Case in point: a new report from data integration giant Fivetran that says 71% of companies find it difficult to access all the data needed to run AI programs, workloads and models.

Working with Vanson Bourne, the company surveyed 550 IT and data science professionals in multiple countries and found gaps in data movement and access across their organizations. The finding is significant as data is vital for model training and implementation. One cannot run a successful AI program without laying a solid foundation for data storage and movement, starting with a data warehouse or lake to automate data ingestion and pre-processing.

“Analytic teams that utilize a modern data stack can more readily extend the value of their data and maximize their investments in AI and data science,” George Fraser, CEO of Fivetran, said in the study.

Data access obstacles

In the survey, almost all of the respondents confirmed that they collect and use data from operational systems on some level. However, 69% said they struggle to access the right information at the right time, while at least 73% claimed to face difficulty extracting, loading and transforming the data and translating it into practical advice and insights for decision-makers.

As a result, even though a large number of organizations (87%) consider AI vital for business survival, they fail to make the most of it. Their broken, manual data processes lead to inaccurate models, eventually resulting in a lack of trust and circling back to humans. The survey respondents claimed that inefficient data processes force them to rely on human-led decision-making 71% of the time. In fact, only 14% of them claimed to have achieved advanced AI maturity — using general-purpose AI to automatically make predictions and business decisions.

On top of that, there’s significant financial impact, with respondents estimating they are losing out on an average of 5% of global annual revenues due to models built using inaccurate or low-quality data.

Talent gets wasted

The challenges associated with data movement, processing and availability also mean that the talent hired to build AI models ends up wasting time on tasks outside of their main job. In the Fivetran survey, the respondents claimed that their data scientists devote 70% of their time on average to just preparing data. As many as 87% of respondents agreed that the data science talent within their organization is not being utilized to its full potential.

According to Fortune Business Insights, the global AI market is projected to grow from $387.45 billion in 2022 to $1,394.30 billion by 2029, with a CAGR of 20.1%