The Chief AI Officer’s role overlaps the role of CIO and Chief Data Officer and without unquestioned ownership of AI, the ability to move ahead in the journey to Everyday AI, a cultural shift where an organisation’s entire workforce operates with AI in mind, is limited, explains Sid Bhatia at Dataiku.
As knuckles whiten and pulses race, the AI rollercoaster seems out of control. The Chief AI Officer is a role that has garnered much attention since AI became a wild stallion in need of a whisperer. While the idea may seem natural and timely, the complexity behind real-world appointments is worth exploring.
Let us start with the basics. The Chief AI Officer’s job description shows much overlap with other technology executives, such as the Chief Information Officer, CIO, and Chief Data Officer, Chief Data Officer. Without absolute, unquestioned ownership of AI, the ability of the Chief AI Officer to effect real purpose in the journey to Everyday AI, an all-embracing cultural shift where an organisation’s entire workforce operates every second with AI in mind is limited.
For the Chief AI Officer to be worth the resources expended in their recruiting, recruitment, onboarding, and ongoing cost of retention, the enterprise must use the resource properly. The Chief AI Officer must be allowed to take the reins of strategy, implementation, and governance so they may guide the organisation towards fulfilling business objectives and, in the case of private companies, gaining competitive advantage.
Where a Chief AI Officer is present, the CIO should be restricted to focusing on broader IT infrastructure, and the Chief Data Officer to looking after data assets. The Chief AI Officer will develop and execute AI programmes. They will align outputs with business goals and drive an Everyday AI culture that embeds ethics, banishes bias, exudes transparency, and delivers data privacy.
In short, the Chief AI Officer is an amalgam of business strategist and risk manager. As part of this hybrid role, Chief AI Officer will lay the groundwork for AI success by devising ways of acquiring and retaining top AI talent. But pushback from CIOs and CDOs is inevitable, given the enormous stake they may already have in AI-related areas.
The CIO may have established the organisation’s overall technology strategy, IT infrastructure, cybersecurity posture, and digital transformation programmes from the ground up with AI an integral part of their portfolio. They may see the Chief AI Officer less as a horse whisperer and more as an unnecessary stable hand.
The Chief Data Officer may have a similar view, given that their role may be considered to extend naturally into AI, including its governance. From the data expert’s perspective, a Chief AI Officer could threaten non-AI data initiatives such as self-service analytics.
The first and most obvious countermeasure to potential power struggles is to define each role to eliminate overlaps. Acknowledge the tight interdependencies between AI, data, and IT, and mandate close collaboration between the Chief AI Officer, CIO, and Chief Data Officer to ensure AI initiatives are supported by all departments and aligned with business goals.
Over time, it is to be hoped that trust and efficiency will emerge from this collaboration. But the Chief AI Officer must lead in AI governance. In any enterprise, a risk to one is a risk to all. Other stakeholders must be encouraged to see the difference between AI and data governance and to recognise the need to support the Chief AI Officer as they take the lead in the former.
But just because the role of AI provider is clearly defined, it does not necessarily follow that responsibilities for Everyday AI should be on the Chief AI Officer’s shoulders alone. Collaborative approaches that include multiple team members from across the organisation could be used to turbocharge AI maturity.
Committees may emerge that either replace the Chief AI Officer or are chaired by the Chief AI Officer but include the CIO, Chief Data Officer, and many other business stakeholders. Committee members would share ownership of AI initiatives, allowing leaders to align strategy and governance on IT, data, and AI before a single project is launched.
This first-things-first approach has proved a winner in general digital transformation programmes. It would be a similar boon to the AI journey, with or without a Chief AI Officer. We have seen so many digitalisation efforts collapse under the weight of departmental silos.
Lessons cannot be salvaged from failures while finger-pointing persists. When siloes are eliminated and everyone takes responsibility for every project, we have achieved joint accountability, a key component of Everyday AI and a strong countermeasure to turf wars.
So, is the wild-stallion wrangler just visiting, or here to stay?
While the Chief AI Officer role may be fulfilled by a cross-functional committee in some organisations, others may see a dedicated executive as a must-have. Either way, the Chief Data Officer and CIO will be critical to the campaign for an AI future, not as lieutenants to the Chief AI Officer, but as fellow generals working together to overcome the complexities of AI integration in a world hungry for its leverage.
Key takeaways
- The Chief AI Officer is a role that has garnered attention since AI became a wild stallion in need of a whisperer.
- The Chief AI Officer’s role overlaps with other technology executives, such as Chief Information Officer, CIO, and Chief Data Officer.
- For the Chief AI Officer to be worth the resources, the enterprise must use this resource properly.
- The Chief AI Officer must be allowed to take the reins of strategy, implementation, and governance so they may guide the organisation.
- Where a Chief AI Officer is present, the CIO should be restricted to focusing on broader IT infrastructure, and Chief Data Officer to looking after data assets.
- The Chief AI Officer will develop and execute AI programmes.
- The Chief AI Officer will align outputs with business goals and drive an Everyday AI culture.
- The Chief AI Officer is an amalgam of a business strategist and risk manager.
- The Chief AI Officer will lay the groundwork for AI success by devising ways of acquiring and retaining top AI talent.
- Pushback from CIOs and CDOs is inevitable, given the enormous stake they may already have in AI-related areas.
- The CIO may have established the organisation’s overall technology strategy, with AI an integral part of their portfolio.
- CIO may see the Chief AI Officer less as a horse whisperer and more as an unnecessary stable hand.
- The Chief Data Officer may have a similar view, given their role may be considered to extend naturally into AI, including its governance.
- From the data expert’s perspective, a Chief AI Officer could threaten non-AI data initiatives such as self-service analytics.
- The first and most obvious countermeasure to potential power struggles is to define each role to eliminate overlaps.
- Acknowledge the tight interdependencies and mandate close collaboration between the Chief AI Officer, CIO, and Chief Data Officer.
- It is hoped trust and efficiency will emerge from this collaboration, but Chief AI Officer must lead in AI governance.
- A risk to one is a risk to all and stakeholders must be encouraged to see the difference between AI and data governance.
- It does not necessarily follow that responsibilities for Everyday AI should be on the Chief AI Officer’s shoulders alone.
- Committees may emerge that either replace Chief AI Officer or are chaired by the Chief AI Officer but include the CIO, Chief Data Officer.
- Committee members would share ownership of AI initiatives, allowing leaders to align strategy and governance, before a single project is launched.
- This first-things-first approach has proved a winner in general digital transformation programmes.
- We have many digitalisation efforts collapse under the weight of departmental silos.
- Lessons cannot be salvaged from failures while finger-pointing persists.
- When everyone takes responsibility for every project, we have achieved joint accountability.
