Research from the MIT Center for Information Systems Research (MIT CISR) shows that organizations with higher AI maturity outperform companies that are still in the early stages of AI adoption. However, many enterprises continue to face challenges in moving from AI pilots to AI at scale.

According to a new MIT CISR research briefing, companies are making progress in AI maturity, with the greatest financial impact occurring when organizations move from building AI capabilities to scaling AI across the business.

Financial Performance Improves as AI Maturity Advances

Last year, MIT CISR researchers introduced a four-stage enterprise AI maturity framework designed to help organizations create value from AI.

The researchers found that enterprises in the first two stages of AI maturity had financial performance below the industry average. In contrast, enterprises in Stages 3 and 4 achieved financial performance well above the industry average.

An update by researchers Stephanie Woerner, Peter Weill, Ina Sebastian, and Evgeny Káganer found that the largest financial gains occur when organizations advance from Stage 2 to Stage 3.

In Stage 2, enterprises focus on developing AI capabilities, including business case development and testing, process simplification and automation, and employee experimentation with AI. In Stage 3, organizations build on these capabilities to scale AI across the business and embed AI into their operations.

Four Challenges Organizations Must Address

Based on interviews with senior executives about their AI initiatives, the researchers identified four challenges that organizations must address to move from Stage 2 to Stage 3.

The first challenge is strategy. Organizations need to align AI investments with strategic goals while ensuring that AI initiatives deliver measurable and scalable value.

The second challenge is systems. Enterprises must build modular and interoperable platforms and data ecosystems that support enterprise-wide intelligence.

The third challenge is synchronization. This involves creating AI-ready people, roles, and teams while redesigning work around AI capabilities.

The fourth challenge is stewardship. Organizations need to embed and monitor compliant, human-centered, and transparent AI practices by design.

Guardian Uses AI to Improve Customer Experience and Efficiency

Guardian Life Insurance is applying AI in three main areas: improving customer experience, increasing efficiency and reducing operating expenses, and improving employee productivity.

To address strategy, Guardian’s data and AI team manages AI strategy and prioritization. The company uses a value-tracking framework that guides initiatives from hypothesis to pilot and then to scale. One pilot project automated the request-for-proposal (RFP) and quoting process, reducing turnaround time from about one week to 24 hours. Guardian plans to scale the project in 2026.

For systems, the company modernized legacy systems and data architecture to support AI scaling. Its chief technology officer reorganized the technology function around products and platforms, supported by small cross-functional teams, microservices, and APIs that enable reuse and faster delivery.

To support synchronization, Guardian is reskilling employees by moving them into AI-focused roles and emphasizing the solution of end-to-end business problems. The company also plans to introduce rotations and training programs to build hybrid business and technical skills.

Regarding stewardship, Guardian has embedded governance through collaboration with risk, legal, and compliance teams. Architecture reviews are conducted through both formal and fast-track review boards to ensure privacy, security, and regulatory requirements are incorporated into new solutions.

Italgas Embeds AI Across Infrastructure Operations

Italgas, Europe’s largest natural gas distributor, is using AI to manage infrastructure, improve efficiency, and enhance safety. The company’s Digital Factory innovation hub supports these efforts through executive sponsorship and cross-functional teams.

For strategy, Italgas prioritizes AI initiatives such as WorkOnSite and DANA. WorkOnSite has accelerated construction projects by 40% and reduced inspections by 80%, while DANA is a generative AI-based network control system. Each minimum viable product sprint is supported by a C-level sponsor to maintain strategic alignment.

For systems, the company has been digitizing assets and processes since 2017. It has developed a cloud-based platform supported by Internet of Things infrastructure, a 300-terabyte data platform, and 23 AI models. Business translators embedded within business units help drive adoption and application of modular components.

To support synchronization, Italgas engaged more than 1,000 employees in innovation initiatives and delivered 30,000 hours of AI and data training in 2024. Through Italgas Academy, employees are mapped to a new digital leadership model designed to build an agile and AI-ready workforce while maintaining continuity.

For stewardship, governance structures include a chief people, innovation, and transformation officer, an AI officer, and a group AI office that oversee AI integration and monitoring. The company balances efficiency improvements with new business opportunities, including the commercialization of WorkOnSite, which generated €3 million in revenue in 2024.

Advancing AI Maturity Requires Organizational Change

The researchers emphasize that progressing through the stages of AI maturity represents a significant organizational transformation. Businesses are likely to encounter both technological complexity and human resistance during this transition.

According to the researchers, successful change requires alignment among the CEO, CIO, chief strategy officer, and head of human resources.

“Now is the time for executive teams to align, commit, and lead the charge toward enterprise-scale AI by developing a playbook for strategy, systems, synchronization, and stewardship,” the researchers write.

PHOTO: FREEPIK

This article was created with AI assistance.

We make every effort to ensure the accuracy of our content, some information may be incorrect or outdated. Please let us know of any corrections at [email protected].