== Co-Authors: Myles Suer and Brian Lett ==
Artificial intelligence (AI) adoption remains uneven across organizations, with overall penetration, regardless of technique, still below 50%. This includes the following AI disciplines: data science and machine learning (DSML), generative AI, and the emerging class of agentic AI. Despite the hype, a majority of organizations are still early in their AI journeys, experimenting selectively rather than deploying AI at scale throughout their core business processes.
For most organizations, the main question about generative and agentic AI is not whether they will be adopted, but rather how soon, and in what ways they will be established in organizations. At the end of 2025, slightly more than half of all organizations reported experimentation with generative and agentic AI—indicative of strong intent, with the main questions being when, how, and where AI starts supporting and enabling business operations.
A significant number of organizations already indicate production use of generative AI (34%) and agentic AI (15%). Both of these adoption levels have more than doubled since 2024. The data on budgets for generative and AI offers further evidence of the inevitability and rapid deployment of these technologies. At the end of 2025, a majority organizations report allocating or reallocating some of their technology budgets to generative AI (72%) and agentic AI (66%).
Data feeds all AI approaches. Data quality plays the primary role in determining whether new applications of these AI approaches meaningfully transform the business or reduce profitability, hamper competitive positioning, and thwart careers. Executives and data leaders who realize data maturity represents the primary constraint to AI effectiveness will also perceive and address data as a strategic asset. They will commit to a concrete roadmap to enable the organization to reach data maturity. In partnership with senior executives, they will communicate and implement a pragmatic, time-bound plan for providing prioritized, industrialized data that will be leveraged throughout the organization—especially in all AI approaches. Without a solid data foundation, AI remains tactical; with it, AI becomes truly strategic.
Organizations can achieve this data industrialization by using a stepped approach that incrementally advances capabilities: Enhancing analytical data infrastructures, improving data quality, implementing and enriching data governance, and eliminating data silos and providing a single source of the truth. By emphasizing data industrialization as a strategic imperative and prioritizing achievable, high-value use cases in all AI approaches, organizations can best position themselves to deliver on the promise of AI.
You do not have permission to access this document. Make sure you are logged in and/or please contact Danielle with further questions.