Data Challenges Emerge as Major Hurdle in AI Adoption and Business Strategies

Why data is the Achilles Heel of AI (and every other business plan)

Companies need to put their data houses in order before moving ahead with generative AI initiatives, warn two new surveys. Excitement is over the top about all the marvels of today's technology -- artificial intelligence, real-time analytics, virtual reality, and connected enterprises, to name a few. However, without the right data, these initiatives are dead in the water. Two new surveys warn that companies still need to put their data houses in order, and as a result, aren't ready to move forward with initiatives such as generative AI (gen AI). There's an uneasy dance going on between data handling and AI development across the business landscape. The challenge is that data remains too much of a risk, rather than an asset in data-driven or AI-based initiatives. Also: Could AI make data science obsolete? While charging headlong into AI and other cutting-edge initiatives, "many organizations don't understand how to value investment in technology and databases -- still viewing them as purely cost centers," said Steve Mitchell, CFO at Redgate Software. "However, there are businesses that have demonstrated the growth and huge value creation opportunity that data -- and the ability to move quickly in harnessing the ever-increasing volume of it -- present. More organizations need and will look for more robust ways to measure the benefit that faster and improved data-focused decision-making can bring -- improved commercial execution, less wasted effort and resource, more satisfied team, and much more." While AI continues to be a priority for IT investment, momentum is slowing due to data dilemmas, a survey of 1,000 IT executives out of Presidio finds. At least 86% report data-related barriers, such as difficulties in gaining meaningful insights and issues with real-time data access. Half of the executives surveyed believe they plunged into gen AI before they were fully prepared, the survey suggests. Among those who have already adopted gen AI, 84% experienced issues with their data sources. "This suggests that readiness isn't just about adopting the technology -- it's about having the right data and infrastructure in place," the survey's authors suggest. There is also hesitation in operationalizing AI. More than nine in ten IT leaders, 92%, report concerns about integrating AI into operations. Also: How Claude's new AI data analysis tool compares to ChatGPT's version (hint: it doesn't) One in five respondents, 20%, caution that AI projects fail due to rushing into implementations too quickly. Another 17% cite data quality issues. This is particularly apparent among healthcare executives, where more than a quarter, 27%, point to hasty adoption as a primary cause of failure. The path to AI and data-driven success is built on governance, and this is currently a struggle for many companies, according to a separate survey of 220 business and IT professionals by Quest Software and Enterprise Strategy Group. The survey finds AI data readiness and operational efficiencies are now top of mind for many executives. Evolving data and governance to an AI-ready state was cited by 33%, making it a top-three bottleneck impacting an organization's data value chain, behind understanding the quality of source data at 38% and tied with the 33% who report challenges with finding, identifying and harvesting data assets. Also: 5 tips for choosing the right AI model for your business The respondents reported that governing the use of AI models and data -- to deliver data mapping, data lineage, and data policies -- is their most difficult management challenge. AI governance topped the list with metadata management -- a key component of AI data readiness -- rising by 21% year over year. Data quality monitoring, data quality remediation, data profiling and quality scoring, and data policies and control rounded out the top challenges with which organizations are currently grappling.

Why data is the Achilles Heel of AI (and most every other business plan)

Companies need to put their data houses in order before moving ahead with generative AI initiatives, warn two new surveys. Excitement is over the top about all the marvels of today's technology -- artificial intelligence, real-time analytics, virtual reality, and connected enterprises, to name a few. However, without the right data, these initiatives are dead in the water. Two new surveys warn that companies still need to put their data houses in order, and as a result, aren't ready to move forward with initiatives such as generative AI (gen AI). There's an uneasy dance going on between data handling and AI development across the business landscape. The challenge is that data remains too much of a risk, rather than an asset in data-driven or AI-based initiatives. Also: Could AI make data science obsolete? While charging headlong into AI and other cutting-edge initiatives, "many organizations don't understand how to value investment in technology and databases -- still viewing them as purely cost centers," said Steve Mitchell, CFO at Redgate Software. "However, there are businesses that have demonstrated the growth and huge value creation opportunity that data -- and the ability to move quickly in harnessing the ever-increasing volume of it -- present. More organizations need and will look for more robust ways to measure the benefit that faster and improved data-focused decision-making can bring -- improved commercial execution, less wasted effort and resource, more satisfied team, and much more." While AI continues to be a priority for IT investment, momentum is slowing due to data dilemmas, a survey of 1,000 IT executives out of Presidio finds. At least 86% report data-related barriers, such as difficulties in gaining meaningful insights and issues with real-time data access. Half of the executives surveyed believe they plunged into gen AI before they were fully prepared, the survey suggests. Among those who have already adopted gen AI, 84% experienced issues with their data sources. "This suggests that readiness isn't just about adopting the technology -- it's about having the right data and infrastructure in place," the survey's authors suggest. There is also a hesitation to operationalize AI. More than nine in ten IT leaders, 92%, report concerns about integrating AI into operations. Also: How Claude's new AI data analysis tool compares to ChatGPT's version (hint: it doesn't) One in five respondents, 20%, caution that AI projects fail due to rushing into implementations too quickly. Another 17% cite data quality issues. This is particularly apparent among healthcare executives, where more than a quarter, 27%, point to hasty adoption as a primary cause of failure. The path to AI and data-driven success is built on governance, and this is currently a struggle for many companies, according to a separate survey of 220 business and IT professionals by Quest Software and Enterprise Strategy Group. The survey finds AI data readiness and operational efficiencies are now top of mind for many executives. Evolving data and governance to an AI-ready state was cited by 33%, making it a top-three bottleneck impacting an organization's data value chain, behind understanding the quality of source data at 38% and tied with the 33% who report challenges with finding, identifying and harvesting data assets. Also: 5 tips for choosing the right AI model for your business The respondents reported that governing the use of AI models and data -- to deliver data mapping, data lineage, and data policies -- is their most difficult management challenge. AI governance topped the list with metadata management -- a key component of AI data readiness -- rising by 21% year over year. Data quality monitoring, data quality remediation, data profiling and quality scoring, and data policies and control rounded out the top challenges with which organizations are currently grappling.