December 10, 2025—The Software Engineering Institute (SEI) today released a white paper describing the forthcoming Artificial Intelligence (AI) Adoption Maturity Model, which aims to help organizations navigate the complexities of implementing this impactful technology. In the frenzy to embrace AI, many organizations are not leveraging its potential for significant benefits. The paper, A Preliminary Report on a Model for Maturing AI Adoption: From Hype to Achieving Repeatable, Predictable Outcomes, defines a maturity approach to incorporating AI into workflows and tech ecosystems: identifying use cases, institutionalizing practices, focusing on the value of investments, and creating a structured roadmap for adoption.

From Hype to Rigor

In partnership with Accenture, the SEI is developing the AI Adoption Maturity Model, which organizations can use to create a roadmap for predictable AI adoption and realize the benefits of AI.

Many AI early adopters have not seen the expected gains in productivity or revenue. While Stanford University found that AI investments, optimism, and accessibility are rising, a recent MIT report suggests that despite investments of $30 billion to $40 billion into generative AI, 95 percent of organizations are realizing no returns. Research from Accenture found that only 8 percent of companies are scaling AI at an enterprise level and embedding the technology into core business strategy to maximize value.

Mismatched expectations, misaligned applications, and poorly executed or untested implementation practices—not the technology itself—often keep organizations from realizing immediate value from a technology investment. For AI to increase efficiency, productivity, and value while conserving resources and lowering overall costs, organizations need to shift their focus from hype-driven experimentation to foundational capabilities and practical, measurable outcomes.

“With AI, supply chain dependencies are wider, and the resource and infrastructure demands can be crushing,” said Ipek Ozkaya, technical director of the SEI’s Engineering Intelligent Software Systems directorate and lead author of the new white paper. “You need to make strategic decisions about what your organization needs, what partnerships to build, and what it can sustain.”

The SEI’s forthcoming maturity model is a framework for assessing the ability of an organization or unit to perform and sustain specific technical practices to achieve its AI adoption goals. The model divides AI-relevant capability areas into eight core dimensions: Organizational Strategy, Workforce and Culture, Workflow Re-engineering, Risk and Governance, Data, Engineering, Operations and Sustainment, and Ecosystem.

An organization’s achievement of the model’s capability areas across each dimension will indicate one of five levels of AI adoption maturity:

1. Exploratory AI: Exploration is the start of transformation.
2. Implemented AI: The organization (or unit) is on the path.
3. Aligned AI: AI workflows are being managed consistently.
4. Scaled AI: AI is used successfully at scale with repeatable results.
5. Future-Ready AI: The organization (or unit) has a record of success with incremental and innovative improvements powered by AI.

“Getting reliable assessments of AI-relevant capabilities and understanding where AI initiatives will provide value are often the initial roadblocks for organizations,” said Rajendra Prasad, group chief executive, Accenture Technology and Chief Technology Officer. “The AI Adoption Maturity Model being developed by the SEI and Accenture will provide a foundation for both issues. With that information, organizations can determine where their AI initiatives are providing value, reinvest saved resources, and institutionalize adoption.”

Leaning Into the Future

AI capabilities such as generative and agentic AI are evolving too rapidly for organizations to adopt effectively without guidance. To institutionalize mature AI adoption, organizations must apply disciplined software engineering practices in-house or manage the external software supply chain effectively. The AI Adoption Maturity Model incorporates the SEI’s latest research in AI and lessons learned from decades of helping organizations with process maturity, software architecture, cybersecurity, Agile software development, and continuous integration and deployment.

The SEI leaned on its lineage of maturity models, from its pioneering Capability Maturity Model (CMM) and CMM Integration (CMMI) to the influential CERT Resilience Maturity Model (CERT-RMM). More recently, it co-developed the Cybersecurity Maturity Model Certification (CMMC).

“The SEI’s work in maturity modeling has provided a critical foundation for building software systems that are safe, predictable, and evolvable,” said Anita Carleton, director of the SEI’s Software Solutions Division. “Now this challenge has reached a new level as fast-moving AI technologies develop, and the SEI is again providing a framework to help organizations make sense of the process and harness AI’s enormous potential.”

To inform the new model, Ozkaya and a team of SEI experts systematically reviewed current AI and other maturity assessment practices, challenges, and needs. Interviews with over two dozen executives and an ongoing survey revealed real organizational challenges and successes with AI. The result, said Ozkaya, will be a new maturity model fitted to today’s biggest technology advancements and flexible enough for the fast-changing AI future. The SEI team is currently piloting the model in advance of an expected April 2026 release.

A New Scale of Technology Adoption

AI stands to consume resources and deepen dependencies more than any previous software technology. Organizations, especially those in highly regulated environments such as finance and healthcare, must target AI adoption in smart ways before re-engineering business processes, workflows, and technology ecosystems. If implemented strategically, however, AI could improve an organization’s productivity and efficiency, lower costs, and speed innovation.

Government agencies will also need a rigorous approach. The Department of War’s recently announced critical technology area of Applied AI is one of many signals that federal programs need a way to adopt AI quickly, responsibly, and effectively.

“Our guiding vision described in the National Agenda for Software Engineering was one in which the current notion of software development was replaced by one where humans and software are trustworthy collaborators that rapidly evolve systems,” said Carleton. “This is happening even faster with the constant evolution of AI technologies. Our AI Adoption Maturity Model builds on the SEI’s past work to ensure AI systems will be as safe, predictable, and evolvable as they are transformational.”

Seeking Pilot Participants

The SEI seeks organizations to participate in its ongoing pilots of the AI Adoption Maturity Model. Those interested can email info@sei.cmu.edu. Organizations can also take an ongoing survey on the challenges and successes in adopting AI.

Download the white paper A Preliminary Report on a Model for Maturing AI Adoption: From Hype to Achieving Repeatable, Predictable Outcomes from the SEI Digital Library. To learn more on this topic, read the SEI Blog post From Hype to Adoption: Guiding Organizations in Their AI Journey.