Global Study Exposes Overconfidence And Strategic Gaps In AI Implementation By Organizations

Nearly half of IT leaders believe their organizations are fully prepared to benefit from artificial intelligence (AI), according to a research report commissioned by Hewlett Packard Enterprise (HPE). However, the report also reveals critical gaps in their strategies that could hinder successful AI outcomes. The survey, titled 'Architect an AI Advantage', gathered responses from over 2,000 IT leaders across 14 countries. It found that businesses are overlooking key areas that are crucial for delivering successful AI outcomes. These areas include low data maturity levels, deficiencies in networking and compute provisioning, and ethics and compliance considerations.

The report highlights a lack of alignment between processes and metrics as a major issue. This misalignment can lead to fragmentation in approach and exacerbate delivery issues. The research also uncovers significant disconnects in strategy and understanding that could impact future return on investment (ROI). Despite the gaps in strategy and understanding, the report shows that the majority of IT leaders are planning to increase their AI spend in the coming year.

One of the key findings of the report is the low data maturity levels among organizations. While data management is considered critical for AI success, only a small percentage of organizations have the capability to run real-time data pushes and pulls. Additionally, less than 60% of respondents said their organizations are capable of handling key stages of data preparation for AI models, such as accessing, storing, processing, and recovering data. This discrepancy not only slows down the AI model creation process but also increases the likelihood of delivering inaccurate insights and a negative ROI.

The survey also reveals a gap in understanding the compute and networking requirements for AI across the entire lifecycle. Although a high percentage of IT leaders believe their network infrastructure and systems are capable of supporting AI traffic and demands, less than half have a full understanding of the demands of various AI workloads. This lack of understanding raises concerns about the accuracy of provisioning for AI requirements.

The report indicates that organizations are failing to connect key areas of business, as over a quarter of IT leaders describe their organization's AI approach as "fragmented." Many organizations have chosen to create separate AI strategies for individual functions, and some are even setting different goals altogether. Additionally, ethics and compliance are being overlooked, despite growing scrutiny from consumers and regulatory bodies. Legal and ethical considerations were deemed the least critical for AI success by IT leaders, and a significant number of organizations are not involving legal teams in AI strategy conversations.

The report also highlights the business risks associated with the lack of AI ethics and compliance. Without proper policies in place, businesses risk exposing their proprietary data and developing models that lack compliance and diversity standards. This can result in negative impacts on brand reputation, loss in sales, and costly fines and legal battles. The quality of AI outcomes is directly related to the quality of the data they ingest, and the low data maturity levels mentioned earlier further contribute to the risk of developing ineffective models.

Dr. Eng Lim Goh, SVP for Data & AI at HPE, emphasizes the importance of a hybrid and modern AI architecture to effectively deliver on the promise of AI. He warns businesses to carefully weigh the balance of being a first mover in AI while fully understanding the gaps across the AI lifecycle. Failure to do so can lead to large capital investments delivering a negative ROI.

The report commissioned by HPE highlights the gaps and challenges that organizations face in realizing the benefits of AI. Despite the growing investments in AI, businesses need to address critical areas such as data maturity, compute and networking requirements, and ethics and compliance considerations. Failure to do so can hinder the success of AI initiatives and result in negative outcomes and wasted investments.