7 Data quality is now a primary barrier to AI adoption In 2024, establishing data standards was framed primarily as a foundational requirement for unlocking AI's potential in the energy sector. The focus was on creating unified data formats and protocols to enable the efficient flow of information across an increasingly complex energy system. This presented data as a critical enabler with the implicit understanding that high-quality, standardized data was a prerequisite for success. Today, the perspective has shifted significantly. The foundational need for good data has now become a critical, top-tier operational barrier. "Data quality and consistency" is now perceived as the second biggest barrier to AI adoption, cited by 45% of leaders, placing it ahead of cost and the lack of skilled talent, and only slightly behind the primary concern of cybersecurity risks. This elevation from a priority action item to a major real-world blocker suggests companies are directly encountering the challenges of working with inadequate data as they look to scale AI deployments. This issue reflects the industry's deepening engagement with AI. As organizations have begun to pilot and integrate, the initial focus on strategic frameworks has given way to a sharp awareness of practical, on-the-ground challenges. Insights from the 2025 survey show that leaders now see "Data utilization/ Accuracy of predictions" as the single most important metric for assessing AI's success in the energy transition, selected by 25% of respondents. The focus has moved from how to get the data to an acute awareness that the quality of that data directly determines the value of their AI investments. As energy companies embrace AI and digital transformation, cybersecurity is essential to protecting critical physical and IT systems. Microsofts Secure Future Initiative (SFI) embeds secure-by-design, secure-by-default, and secure-operations principles across product development, deployment, and operations mobilizing thousands of engineers to strengthen identity protections and accelerate vulnerability remediation at scale. Complementing these efforts, AI-powered tools like Microsoft Sentinel and Security Copilot analyze trillions of threat signals, deliver contextual insights, and automate remediation to counter sophisticated attacks, including those using generative AI. Together, SFI and AI-driven security capabilities provide the resilience energy companies need to innovate confidently without compromising critical infrastructure. Unlocking real-time value with a unified data platform Microsoft Fabric and Azure Arc create a unified enterprise data foundation that simplifies the integration and scaling of AI across cloud, edge, and on-prem environments. Fabric consolidates data into an open lake for analytics and AI readiness, while Azure Arc extends governance, security and policy wherever data resides. Together, they unlock real-time edge-to-cloud intelligence for faster, smarter decisions. Chevrons Facilities of the Future initiative demonstrates this using Azure IoT Operations to enable remote operations, streamline performance monitoring, anomaly detection, and proactive response to changing conditions. 18 Transforming energy workflows with open data standards Azure Data Manager for Energy is a fully managed, enterprise-grade platform service aligned with the OSDU Technical Standard replacing custom integrations with standardized data products and protocols for secure, scalable, and interoperable data management and workflows. Combined with Microsoft Fabric and Microsoft OneLake, it makes data accessible and AI-ready, automating interpretation and insights. Energy companies are accelerating ingestion to insight, decision-making, and innovation in areas such as carbon capture and storage, reservoir modelling, and operational planning, driving faster time to value. 19 Building a trusted foundation for energy innovation and resilience Cybersecurity and data quality rank higher than cost as greatest barriers to AI adoption Figure 6 Biggest Barrier to AI Adoption Cybersecurity risks 49% Data quality and consistency 45% Cost of implementation 40% Lack of skilled talent 39% Regulatory risk 35% Lost of human oversight 32% Reputional risk 14% AIs potential hinges on data- its quality, accessibility, and scale. As energy systems become more intelligent and interconnected, the ability to unify and analyze vast datasets in real time will define success. The convergence of AI and energy isnt just about automation; its about unlocking insights that drive sustainability, resilience, and innovation at scale. Jake Loosararian CEO, Gecko Robotics 2025 SURVEY INSIGHT Key Shifts from 2024 to 2025 The call to "Establish data standards and protocols" has transformed into a present- day obstacle, with "Data quality and consistency" now ranked as the second biggest barrier to AI adoption by 45% of leaders, surpassing even the cost of implementation. The conversation has moved beyond data infrastructure to business outcomes. "Data utilization/Accuracy of predictions" is now the most important metric for assessing the success of AI in the energy transition. The energy system of the future will be far more complex, with a dramatic increase in smart meters, sensors, and connected devices. Watch For in 2026 Increased investment in data governance with organizations launching comprehensive data cleansing and governance initiatives as a prerequisite for major AI projects. The emergence of specialized roles focused on "data curation for AI" will likely appear within energy companies, tasked with ensuring data is clean, consistent, and fit-for- purpose. A renewed push for industry standards with the earlier call for pre-competitive data standards gaining urgent momentum. 18 Chevron plans facilities of the future with Azure IoT Operations, https://www.microsoft.com/en/customers/story/22849-chevron-iot-operations 19 Azure Data Manager for Energy, https://azure.microsoft.com/en-us/products/data-manager-for-energy 35 Powering Possible 2025 34