How should leaders govern AI systems they cannot fully control?
In July 2026, KPMG answered the question many executives are beginning to confront: how should leaders govern AI systems they cannot fully control? KPMG’s position is that organizations should deploy AI only to the extent they are confident they can control it, putting safeguards and governance mechanisms in place before expanding its use.
How should leaders govern AI systems they cannot fully control?
This question sits at the center of the conversation around artificial intelligence adoption. During Cybersecurity in the Age of AI: The Arms Race Has Already Started at the 2026 KPMG Tech & Innovation Symposium, the discussion highlighted a reality many organizations are already experiencing: AI systems are becoming more capable, more autonomous, and more deeply embedded in business operations. As adoption accelerates, leaders must determine how to govern technologies whose capabilities continue to evolve.
The challenge is not simply technical. Executives are balancing competing priorities. They want to capture the benefits of AI while maintaining confidence that systems are operating as intended, supporting business objectives, and remaining aligned with organizational values and risk tolerances. The result is a growing focus on governance as a business issue rather than solely a technology issue.
Why It’s Harder Than It Looks
As AI systems become more capable and autonomous, leaders face a new governance challenge: balancing innovation with appropriate safeguards. Organizations can benefit significantly from AI, but they also need confidence that the systems they deploy operate within defined boundaries and business expectations. Trust, governance, and risk management therefore become essential to scaling AI adoption responsibly.
Leaders also face pressure from multiple directions. Employees want access to increasingly powerful tools. Customers expect better experiences enabled by AI. Investors want evidence that organizations are taking advantage of emerging technologies. At the same time, boards and regulators are paying closer attention to risk, accountability, and oversight. Governance therefore becomes an exercise in balancing innovation with responsible deployment.
The Evidence
1
2
3
4
KPMG’s Answer
Leaders should begin with a straightforward question: are they confident they can control how the AI system will be used within their organization?
If the answer is no, deployment should be limited until appropriate safeguards, governance mechanisms, and oversight are in place. As AI capabilities continue to advance, organizations need confidence that systems are operating within defined boundaries before expanding adoption. Effective governance begins by defining acceptable outcomes, establishing clear accountability, and creating mechanisms that provide visibility into how AI systems are behaving.
The discussion highlighted the importance of safeguards and alignment. Organizations should define acceptable boundaries for AI use, understand the risks associated with deployment, and ensure governance processes provide visibility into those risks. Effective governance helps organizations adopt AI with greater confidence while maintaining trust.
The pace of AI advancement underscores the importance of regularly evaluating how governance, safeguards, and risk management practices support the organization's AI objectives. Speakers noted that AI capabilities are evolving rapidly, requiring leaders to remain engaged in how these technologies are adopted and managed.
Organizations that delay establishing governance and safeguards may find it more difficult to scale AI adoption with confidence. As AI becomes more deeply embedded across business operations, trust, oversight, and accountability will remain important considerations for leadership teams. As AI becomes more deeply integrated into critical business processes, clear governance will become increasingly important to sustaining trust and confidence.
Establish safeguards before expanding deployment. Organizations should ensure they are comfortable they can control how AI systems are used before scaling adoption.
Balance innovation with risk management. Organizations should pursue AI opportunities while maintaining appropriate guardrails, trust, and governance mechanisms.
Explore more
Get in touch
Start the conversation
Connect with our team today to learn how we can help you realize the full potential of GenAI.
