Who Actually Owns AI Governance?

The Shifting Landscape of AI Governance: Who’s in Charge?

The advertising industry has long been grappling with governance questions, from navigating complex privacy rules to ensuring data minimization and consent. The addition of artificial intelligence (AI) to the mix has introduced another layer of complexity, leaving many to wonder: who actually owns AI governance?

Advertisers have been dealing with governance questions for years: privacy rules, data minimization, consent, dark patterns, kids – you name it. But now AI is adding another layer of complexity, and a lot of that work is landing on the same people who already juggle privacy, security and data governance. There isn’t yet a clear line between privacy and AI governance, which makes basic questions like “Who owns this?” and “What does good look like?” surprisingly hard to answer.

The Current State of AI Governance

According to a recent report by the International Association of Privacy Professionals (IAPP), 48% of companies say they lack sufficient budget and resources to invest in governance professionals, while 67% say the primary responsibility for AI governance rests with the privacy function. These numbers point to an AI governance role that’s still being defined inside most organizations.

“There isn’t a consistent model yet,” said Ashley Casovan, managing director of the IAPP’s AI Governance Center. “Our research is survey-based and we’re talking to privacy professionals, of course, which introduces a little bias, but even with that caveat, it’s clear that privacy teams are getting pulled into AI governance.”

The Evolving Role of Privacy Teams in AI Governance

So, how are companies structuring their AI governance right now? The answer is: it looks very different from one organization to the next. In some places, AI governance is added onto what privacy people are already doing. In others, the job has evolved so much that it’s essentially a new role where this person is focused almost entirely on AI governance and someone else has taken over the privacy function.

It’s not just privacy, though. For example, cybersecurity and data governance teams are also playing a role in AI governance. The blurring of lines between these functions requires a rethinking of traditional governance structures.

As Casovan noted, the involvement of multiple functions in AI governance may lead to a more holistic approach, but also creates challenges for coordination and accountability. For instance, who should be responsible for ensuring that AI systems are transparent and explainable? Should it be the privacy team, the data governance team, or someone else entirely?

Insights and Implications

The lack of clear ownership and structure for AI governance may hinder effective implementation and enforcement. Without a clear understanding of who is responsible for AI governance, organizations may struggle to ensure compliance with emerging regulations and industry standards.

Moreover, the evolving role of privacy teams in AI governance raises questions about the skills and expertise required for this work. Do privacy professionals need to develop new skills to effectively govern AI systems? Or will new roles and teams emerge to focus specifically on AI governance?

Case Studies: Different Approaches to AI Governance

Some companies are taking a ** decentralized approach to AI governance**, where multiple teams and functions are responsible for different aspects of AI governance. For example, a company may have a central AI governance team that sets overall strategy and policy, while individual business units are responsible for implementing AI governance in their specific areas.

Others are taking a more centralized approach, where a single team or function is responsible for AI governance across the organization. For instance, a company may have a dedicated AI governance team that works closely with privacy and data governance teams to ensure compliance and effectiveness.

Conclusion

As AI governance continues to evolve, organizations will need to adapt and clarify their approaches to ensure effective governance and compliance. The ownership and structure of AI governance is still unclear, but one thing is certain: it will require a collaborative effort from multiple functions and teams.

In the words of Casovan, “the key is to find a balance between innovation and governance”. By working together and sharing expertise, organizations can ensure that AI governance is effective, efficient, and aligned with business objectives.

The future of AI governance is still being written, but one thing is clear: it will be shaped by the collective efforts of privacy professionals, data governance teams, cybersecurity experts, and business leaders. As the landscape continues to evolve, one question remains: who will ultimately own AI governance? Only time will tell.