Organizations exploring the use of generative AI could encounter significant roadblocks unless efforts are made to improve data maturity, experts have warned.
Data maturity, which refers to the capacity of an organization to collect, store, and use their data effectively, has grown in importance amid a heightened focus on generative AI development in recent months, research shows.
The acceleration of generative AI applications for business means many firms are now augmented or overhauling their strategies to accommodate for ongoing AI projects.
Despite this, senior industry leaders believe that there is still a long way to go before the majority of organizations reach a level of maturity that matches their AI ambitions, according to Accenture CEO Julie Sweet.
Speaking to the Financial Times this week, Sweet said inadequate data capabilities are constraining most businesses, who lack the infrastructure to deploy generative AI at scale.
“The thing that is going to hold it back, though, is . . . most companies do not have mature data capabilities and if you can’t use your data, you can’t use AI,” she told the publication. “That said, in three to five years we expect this to be a big part of our business.”
Speaking to ITPro, Waseem Ali, CEO at data consultancy Rockborne, echoed her thoughts on the matter, suggesting that the benefits organizations can derive from AI are contingent on feeding it good quality data, which needs to be managed appropriately.
“Ultimately, insights from AI are only as good as the data that is put in,” he said.
“So, as mundane as it might be, companies must first establish comprehensive data governance strategies – ensuring data is not only being collected and stored properly but also that it is of a high enough quality to be fed into AI software.”
Ali added that the fierce interest in generative AI over the last year has prompted many business leaders to rush into things without a concise strategy, which long-term will create significant problems.
"The reality is that many businesses want the benefits of AI but don't actually know which business goals AI will help them to achieve. Leaders need to consider AI as they would any other technology–asking themselves questions such as: 'Will this actually help me to achieve X?' and then working backwards to determine what processes would need to be in place to make it possible.”
Most businesses still lack a data strategy
The recently published Data Maturity Index from Carruthers and Jackson highlighted the scale of the data strategy problems facing businesses at the moment.
The Data Maturity Index surveyed data leaders around the world, with 60% of respondents stating their organization is either in the early stages of considering AI or have not yet adapted to meet the technology’s challenges.
Fundamental to an organization’s ability to deploy AI systems will hinge around having a robust data strategy, the study found, yet just 27% of respondents reported their organization had none at all.
Although this marks a slight improvement on the 29% figure recorded in 2022, the lack of progress is concerning, according to Dr Ashley Cairns, delivery director at Carruthers and Jackson.
“For me, the negligible change in the number of organizations with a Data Strategy is one of the most concerning stats from this year’s report,” she said.
“It feels like organizations are still very reactive in their approach to data which, given the sizable investments often needed and the scale to which data underpins an organization's ability to deliver its goals, feeds into a perception that “data” isn’t delivering enough value.”
The report found over half (56%) of respondents described their organization’s data policies as either ‘clunky’ or ‘non-existent’, reflecting no change from 2022’s findings.
Ali said that to address insufficient data infrastructure currently limiting companies, tech leaders must focus on ensuring knowledge is shared and that cross-functional collaboration is improved moving forward.
"Data and tech leaders have more than enough knowledge to overcome challenges, but too often we operate in silos, only sharing information after an incident such as a data breach. Part of this is a matter of pride; not wanting to admit that they do not wholly understand new technologies such as AI. But where possible, companies need to avoid the temptation to let ego get in the way, if not for their own benefit, then for the sake of the state of data security everywhere.”
“Every industry will be facing the same dilemmas and issues, so why not bring everyone around the table to proactively devise frameworks and guide rails that work for them? Instead of waiting around for the slow beast of government policy to catch up, leaders should be putting the wheels in motion for their own regulation, now.”
