Yana Iskayeva/Getty Pictures
Observe ZDNET: Add us as a most well-liked supply on Google.
ZDNET’s key takeaways
- Whereas many corporations appoint CAIOs, others aren’t so positive.
- Insurance coverage specialist Howden has a director of AI productiveness.
- This skilled ensures collaboration and efficient asset exploitation.
There’s plenty of debate about who ought to be answerable for guaranteeing the enterprise makes probably the most out of generative AI. Some consultants recommend the CIO ought to oversee this important function, whereas others consider the duty ought to lie with a chief knowledge officer.
Past these present roles, different consultants champion the chief AI officer (CAIO), a newcomer to the C-suite who oversees key concerns, together with governance, safety, and identification of potential use instances.
Additionally: 5 methods you may cease testing AI and begin scaling it responsibly in 2026
ZDNET reported final 12 months that 60% of corporations have already got a CAIO, and one other 26% are planning to make an appointment this 12 months.
Consultants argue that the rise of the CAIO illustrates AI’s important function in fashionable enterprise. But not everybody believes a devoted CAIO is essentially the perfect answer to the challenges related to implementing AI.
Final 12 months, Kirsty Roth, chief operations and expertise officer at Thomson Reuters, informed me her group does not have a CAIO. “No, we’re not large on these sorts of issues,” she mentioned, suggesting the inherent function of AI in her agency’s processes meant rising expertise have to be thought of by all workers slightly than being seen in isolation.
Nonetheless, importantly, Roth additionally acknowledged that the ever-increasing significance of AI to enterprise operations means somebody senior should guarantee calls for for rising tech are managed successfully: “I suppose when you have no person enjoying these roles, you then most likely want to consider who you’ve got obtained or the way you increase your present managers.”
Additionally: 3 good methods enterprise leaders can construct profitable AI methods – earlier than it is too late
That sentiment resonated with Barry Panayi, group chief knowledge officer at insurance coverage agency Howden, who informed ZDNET that his firm has created a devoted function for a director of AI productiveness, a specialist who sits between Panayi’s knowledge group and the IT division, creating a powerful collaborative interface with the remainder of the enterprise.
So, what does this specialist do? Listed below are three the explanation why your corporation wants a director of AI productiveness.
1. Connecting the whole lot
Many individuals throughout different enterprise models are confused concerning the totally different roles of expertise and knowledge groups. When Panayi joined Howden in August final 12 months, he determined to move off that difficulty on the move.
Barry Panayi: “These instruments are new sufficient that we do want folks to assist with adoption.”
Howden
“The CTO and I sit subsequent to one another, and we mentioned very early on, let’s provide you with a transparent line on this, as a result of if we do not agree, it is not going to be simple. And we each thought the identical factor,” he mentioned, referring to their answer to this intractable problem.
“If you happen to’re shopping for a device and persons are simply utilizing it, then tech ought to personal and run that, as a result of it wants to sit down on their platforms and be plumbed in correctly,” he mentioned.
“There is a bit on our knowledge facet, which is constructing bespoke fashions, like for machine studying. After which there is a bit within the center, the place you do each, so that you would possibly use the ChatGPT API to do the LLM processing for one thing, after which we are going to write code on prime.”
Additionally: 8 pressing updates your IT playbook must survive the AI period
Panayi referred to this cut up between IT and knowledge as “a construct versus purchase choice, with a sliver within the center.” That sliver is the assembly level the place the director of AI productiveness performs an important function.
“I feel corporations are lacking a trick in the event that they’ve not obtained somebody guaranteeing that persons are utilizing issues like Copilot and so forth. These instruments are new sufficient that we do want folks to assist with adoption,” he mentioned.
“And in the intervening time, I do not assume we will assume the narrative is right that individuals utilizing AI at house to assist them guide holidays is identical as the way it may also help them be extra productive at work.”
2. Making certain property are exploited
Panayi mentioned the function of Howden’s director of AI productiveness is to make sure everybody throughout the enterprise makes use of enterprise-grade gen AI providers successfully.
“I see it as a means of sweating expertise and instruments. Their instant concern is ensuring our enterprise is doing its utmost to make use of the instruments that we’re paying for, and Copilot, ChatGPT, and Anthropic are our fundamental three enterprise licenses,” he mentioned.
Additionally: 10 methods AI can inflict unprecedented harm in 2026
“We notice that, at this second, Copilot is nice should you’re in Workplace and also you want AI to summarize issues. Then you definitely’ve obtained Claude, which, should you’re an engineer or in finance, can provide you detailed data. After which there’s ChatGPT, who’s your sort of mind for rent.”
The corporate makes use of different instruments and fashions for particular initiatives. Panayi described overseeing the efficient use of those enterprise-grade and spot options as a job and a half, on condition that Howden employs greater than 20,000 folks in 55 international locations.
“It is like he is a magician, exhibiting individuals who should cope with hundreds of pages of stuff, the way to get the solutions they want rapidly,” he mentioned, outlining how the director of productiveness highlights the advantages of gen AI to the agency’s brokers.
“These persons are not on the pc all day. They’re out out there, speaking and making selections.”
Additionally: How you can truly use AI in a small enterprise: 10 classes from the trenches
Panayi mentioned the efficient exploitation of rising expertise requires a nuanced strategy, which the director of AI productiveness offers by outlining how AI can be utilized safely, securely, and successfully.
“He is obtained so many examples of individuals saying issues like a activity that took them per week to do can now be completed in 20 minutes, with an agent that runs the operation each Monday,” he mentioned.
3. Specializing in aggressive benefit
The director of AI productiveness collaborates with Howden’s IT staff to make sure workers perceive which instruments are prepared and obtainable for work duties.
“Getting everybody utilizing these instruments shouldn’t be an information factor; it is a tech factor. That strategy takes away a ton of demand from the info staff, as a result of managing the demand for gen AI shouldn’t be on my plate, fortunately,” he mentioned.”
Additionally: Nervous concerning the job market? 5 methods to face out within the age of AI
“I am not on the market pushing Copilot and different fashions — that’s an IT tooling choice, which I feel sounds apparent while you say it. However I’ve seen too many knowledge groups drown as a result of they’re making an attempt to handle the group’s use of Copilot.”
Panayi mentioned the director of AI productiveness’s skill to get everybody utilizing enterprise-grade gen AI instruments means his knowledge staff can work on initiatives that produce probably the most bang for the enterprise’s buck.
“All that gen AI effort is nice and highly effective, and creates plenty of productiveness, however I focus extra on the machine studying finish of AI, as a result of I feel there’s unbelievable energy there,” he mentioned.
In an age when many organizations and workers have entry to related off-the-shelf fashions and brokers, Panayi mentioned your aggressive benefit comes from exploiting proprietary knowledge and fashions.
Additionally: 3 methods anybody can begin utilizing AI at work as we speak – safely
“We’re assessing danger, the potential influence, and deciding what the value of our merchandise ought to be,” he mentioned.
“That is a numbers science sport, in order that we can provide the dealer situations to use their data. It is about supercharging the brokers or underwriters with all new insights and concepts, after which they will assemble the product.”
