How Software Boards Are Governing AI

AI is advancing at a pace few organizations have experienced before. Software companies are investing heavily, experimenting across virtually every function, and rethinking how work gets done. Yet for all the activity, many are still working to translate AI adoption into measurable business outcomes. 

McKinsey’s 2025 “The State of AI” survey found that while AI adoption is now widespread, only 39% of organizations report EBIT impact at the enterprise level. For boards, this gap presents a critical challenge. The question is no longer whether companies should invest in AI. It is how to ensure those investments create lasting value.

This topic took center stage during a recent roundtable hosted by Egon Zehnder in partnership with Boardwave, bringing together board Chairs, CEOs and non-executive directors from software companies. While perspectives varied, one conclusion was clear: AI is reshaping what effective governance looks like.

“Companies have never experienced this level of velocity with AI,” one participant observed. 

The speed at which AI capabilities, costs and use cases are evolving is challenging traditional governance models. As a result, AI can no longer be treated as an occasional board topic. It is increasingly appearing in discussions about AI now surfaces across strategy, product development, talent, customer experience and risk discussions. Boards are expected to stay close to these developments while maintaining the right level of oversight.

Many participants argued that governance must become more dynamic. More frequent engagement, metrics, and ongoing education are becoming essential. Several also noted that many boards are still developing the expertise needed for effective AI oversight. Lack of familiarity with emerging AI concepts were cited as recurring challenges, reinforcing the need for accelerated learning and stronger board leadership on AI-related matters.

The software sector has embraced AI experimentation. The harder challenge is turning experimentation into meaningful impact.

As one CEO put it: “What companies are missing is not the use of AI, but its impact.” 

Boards play an essential role in closing this gap. That means ensuring that AI initiatives are anchored in business priorities, supported by clear ownership and measured against concrete outcomes. It also requires discipline in decision-making, including clarity on where to invest, where to scale and where to stop. 

Participants stressed the importance of linking AI directly to value creation. Progress becomes meaningful when organizations can demonstrate how AI contributes to growth, efficiency or competitive advantage. One Chair described using dedicated AI dashboards to monitor indicators such as costs, revenue generation, productivity improvements, customer value creation, and risk exposure as an example to help inform these board discussions.

Participants held differing views on whether boards should introduce dedicated technology or AI committees. Some argued that specialized committees can provide deeper expertise and focus. Others cautioned that isolating AI discussions risks reducing engagement from the broader board on an issue that is increasingly central to company strategy.

While opinions differed on structure, participants were largely aligned on a key point: accountability.

Regardless of how governance is structured, organizations benefit from clear ownership, regular visibility and structured review of AI initiatives. Boards mechanisms that allow them to challenge assumptions, monitor progress and ensure leadership remains focused on outcomes.

The effectiveness of governance will ultimately depend less on where AI discussions take place and more on whether accountability is clearly established.

AI transformation cannot be delegated solely to technology teams.

Participants repeatedly emphasized the importance of the alignment between the CEO and the board, particularly as organizations make decisions that will shape future competitiveness and growth.

The CEO drives execution, while the board provides direction, guidance, challenge and perspective. Together, they play a central role in ensuring that AI initiatives remain connected to broader strategic priorities.

This partnership becomes particularly important when difficult decisions arise around investment levels, organizational change, talent priorities and risk tolerance. Strong alignment allows organizations to move decisively while maintaining appropriate oversight.

While leadership sets direction, innovation often emerges across the organization. 

Participants highlighted the importance of creating space for employees to experiment, learn and identify meaningful use cases. Organizations that generate the most value are often those that combine clear strategic priorities with broad participation.

Building these capabilities requires investment in learning and development, exposure to AI tools and a culture that rewards curiosity, adaptability, and continuous learning. As one participant noted, “What will make the difference is not the tool, but the people.” An individual’s ability to work effectively with AI and embrace ongoing technological change may soon become an important consideration in hiring and leadership decisions.

External ecosystems also play a role. Partnerships with universities and research institutions can expand access to expertise and support the development of future talent. These approaches are more widely used in some markets, such as the United States, and offer relevant lessons for companies looking to accelerate capability building.

AI is introducing a new set of considerations for boards, including cybersecurity, data governance, sovereignty and cost management. At the same time, the opportunity to create value remains significant.

Boards are expected to guide organizations through both dimensions. This involves maintaining visibility on emerging risks while supporting continued progress. Clear governance, strong oversight and a focus on long-term outcomes help organizations move forward with confidence.

Participants also noted that boards need to strike the right balance between avoiding excessive caution and addressing genuine risks, particularly around sovereignty, data governance and dependency on external AI providers. The discussion suggested that risk oversight should not slow down adoption unnecessarily but should help organizations make more informed and accountable decisions.

The discussion underscored a broader shift in how boards create value. AI is raising the expectations around board engagement, adaptability, and speed. Traditional governance approaches may no longer be sufficient in an environment where capabilities and business models are evolving so rapidly.

The boards that will have the greatest impact are likely to be those that stay close to the business, focus on outcomes and continuously build their own understanding of AI. 

Ultimately, the roundtable suggested that boards will need to become more adaptive themselves. In an environment where AI capabilities, costs and use cases are evolving rapidly, the most effective boards will be those able to challenge, learn and recalibrate continuously alongside management.