A study by Cloudera shows that 88 percent of companies use AI to some extent. But a lack of infrastructure and skills prevents them from taking full advantage of AI.

According to a study by Cloudera, many companies still struggle with outdated data infrastructure and a lack of skilled workers, which hinders AI implementation. Cloudera surveyed 600 IT managers about the challenges and obstacles to business AI applications. Although 88 percent of companies already use AI to some extent, they often lack the necessary infrastructure and skills to fully benefit from AI.

AI barriers

AI is gaining traction around the world due to its ability to optimize business processes and accelerate innovation. However, not every company is able to fully leverage the benefits of AI. The biggest barriers to AI adoption are security and compliance risks (74%), a lack of trained staff (38%) and the high cost of AI tools (26%).

Trustworthy data is essential for AI applications. Although 94 percent of respondents say they trust their data, 55 percent struggle to manage data from different sources. Challenges include conflicting data sets (49%), difficulty organizing data across platforms (36%), and data overload (35%). These issues point to the lack of a modern data architecture.

AI applications

AI is currently being used to automate IT processes, create chatbots and improve analytics. The most common applications are:

• Improving the customer experience (60%): AI is used for security and fraud detection (59%), customer service automation (58%), predictive customer service (57%) and chatbot support (55%).

• Increasing operational efficiency (57%): AI is being integrated into numerous business processes, including customer service and marketing.

• Accelerated analysis (51%): Greater access to analytics leads to more informed decisions, providing a competitive advantage. Nearly 80 percent of respondents say their company uses all available data to make smarter business decisions.

“For many organizations, the quality of their data is suboptimal, spread across different infrastructures and poorly documented,” said Abhas Ricky, chief strategy officer at Cloudera. “Managing data wherever it lives is critical to AI adoption. Organizations are beginning to see the benefits of applying AI models to their data, rather than the other way around.”