AI Directors: No Longer a Nice-to-Have, but a Must-Have for Competitiveness

Many fear that artificial intelligence (AI) will take their jobs, but in some companies, thanks to this technology, new job positions (and very well-paid ones) have already started to open up. There is hardly anyone who hasn’t started playing with artificial intelligence tools and using them in their daily work, from writers of all kinds to finance professionals and IT specialists, which has created a need for someone to lead these completely new business processes. Thus, AI directors, or chief artificial intelligence officers (CAIO), have emerged on the business scene, even in Croatia.

Darko Špoljarić, a software developer with decades of experience and the former lead engineer at the IT company Notch, was appointed director of the Artificial Intelligence Department in July. His role is to lead and assist employees and clients in learning to use AI to increase productivity. This, among other things, means testing and deeply understanding various AI models, their capabilities, and discerning situations where it makes sense to develop new AI and where it does not.

– More specifically, I coordinate more than twenty individuals who actively work on AI research projects as part of our internal initiative Devcademy, as well as for our clients. They come from various profiles, including software engineers, MLOps engineers, AI engineers, data engineers, prompt engineers, and foundation model experts – explains Špoljarić, who views his department as a ‘small, fast, and flexible startup‘.

In smaller IT companies, the AI director usually takes care of developing employee competencies, leads experimental projects, and supports clients in deciding when and for what to use AI. In corporations, however, AI directors develop ethical frameworks, manage risks, and plan AI transformations of business processes. In Big Tech, for example, AI has long been recognized as an important area of business. Google, Meta, Microsoft, and Apple have entire departments and even higher-level AI directors, those at the C-level. We increasingly see them in banks, insurance companies, and even pharmaceutical companies.

Strategists in Corporations

Croatia is no exception. For instance, Wiener Insurance in Croatia is currently hiring an AI expert who will focus exclusively on the development and implementation of AI solutions. Board member of Wiener, Božo Šaravanja, emphasizes that they want someone in that position who understands both the technological and business dimensions, can connect professional sectors, IT, and management, and drive change.

– The role of the AI transformation leader will be to identify where artificial intelligence can bring the most value to our business and clients, then set priorities and lead AI initiatives from the first idea to concrete application. Besides specific projects, this person will be responsible for creating an internal AI strategy, collaborating with our development and analytical teams, monitoring EU regulations, and connecting with external experts – lists Šaravanja.

Their ideal candidate for this position understands how AI works, what its advantages and limitations are, and how to meaningfully apply it in everyday business situations. Communication skills are also very important to them because this role, as Šaravanja predicts, ‘will be a bridge between IT, the legal department, professional teams, and external partners’. In the future, as part of digital transformation, they envision a whole AI team that will develop solutions for automating customer support, optimizing internal processes, advanced analytics, fraud detection, and supporting sales agents.

Wiener is just planning to form an AI team, while A1 Croatia has had its Center of Excellence for Data and Artificial Intelligence for some time. Its leader, Andrea Pirša Ilić, emphasizes that leading an AI department in telecom, besides technical aspects, deep understanding of business challenges, and team building, includes ensuring high standards of security, ethics, and transparency in the use of artificial intelligence in the company.

– My AI team consists of fifteen colleagues, machine learning engineers, and data engineers. We collaborate intensively with other departments in multidisciplinary teams. An example of this is the CS squad , which includes members of my AI team as well as customer service experts. Together, we work on developing solutions specific to that business area, ensuring that AI projects align with real processes and the needs of the organization – explains Pirša Ilić, who believes that it is crucial today to apply AI where there is a real business problem that it can solve or ease.

– The point is not to jump on the AI bandwagon at any cost, but to use it responsibly and strategically where it is truly justified – she emphasizes.

AI Washing

In this, lies one of the biggest problems in introducing new technologies into organizations. Many who rushed in without a broader strategy and clear goals may be disappointed when the expected returns on investments in artificial intelligence do not materialize. Does every company need to invest in it and have an AI director just because others do? Ive Botunac, vice president of the Croatian Association for Artificial Intelligence CroAI, believes not. He also notes that there should be a clear vision of how AI contributes to business goals, rather than just following trends.

– Not every company needs to have its own AI department because it requires significant expenditures and engagement, from the hiring process through onboarding to the fact that AI experts are often paid more than programmers in other niches. The advantages of having your own AI department include faster product development, better control over intellectual property, and the ability to create unique competitive advantages. The potential challenges are significant: high costs, difficulties in finding quality personnel, and the constant need for team education. There is also a risk of AI washing, introducing AI just for the trend, without clear business value – emphasizes Botunac.

He recommends smaller companies start with consulting services or partnerships with AI experts, gradually building their own teams when they reach critical mass and clearly define their AI strategy.

For instance, Infinum is aware of this, which is why they currently do not plan to hire an AI director. However, if the need arises for expertise in a very specific area, they are open to collaborating with external consultants, says Ivan Đikić, vice president of engineering at the company. When asked whether every company, or at least IT companies, should have an AI director, he responds that it is difficult to give a universal answer as it all depends on their strategic goals and area of work.

– If we are talking about implementing AI tools to improve internal processes and efficiency, it is often more effective to start with a smaller agile team composed of members with different expertise. Such a team can define the rules for using AI tools and collaborate with other departments. At some point, that team may grow into a separate department, but our experience shows that in rapid technological changes, flexibility is more important than strict hierarchy. A too-rigid structure can slow progress and make it difficult to adapt to new opportunities. The most important thing in this whole story is that every IT company has a clear strategy and plan for artificial intelligence, regardless of whether it is about implementing AI tools in daily work, developing their own AI solutions, or offering such services to the market – explains Đikić.

Logical Development

However, Botunac, who previously led an AI department in a company himself, notes that the current trend of appointing AI directors and establishing AI teams in Croatian companies is ‘a completely natural and logical development that reflects the state of the market’.

– Croatian companies recognize that AI is no longer a nice-to-have but a must-have for maintaining competitiveness. Many new jobs have emerged, from AI engineers to AI researchers to AI product managers. AI positions are predominantly filled by experienced programmers looking to make a career change. It is easier for them because they already have a technical foundation, understand programming, system architecture, and can relatively quickly adapt to AI tools and frameworks. Here we are not talking about machine learning experts; that is a significantly more demanding area that requires a deep understanding of mathematics and statistics. AI engineers mainly use existing tools and APIs to apply AI functionalities without necessarily understanding how the underlying models work. We also see more and more young people directly orienting themselves towards AI already in college, especially due to the emergence of AI programming assistants like Cursor or GitHub Copilot. The term vibe development, which refers to an approach where you rely on AI tools to solve problems, is even popular – explains Botunac, emphasizing that AI directors must ensure compliance with regulations such as the European AI Act.

Indeed, the new European regulation brings strict requirements related to model transparency, documentation, data used, oversight of high-risk systems, and bans on certain practices, and it also foresees fines of up to seven percent of global revenue. This means that AI directors will also have to care about compliance to avoid unwanted consequences for the company’s budget and reputation.

Nothing Without Results

Therefore, if an AI director can contribute to achieving the company’s goals with their technical and business skills or help a startup scale its business faster, their appointment is certainly desirable. However, merely bringing in a new leader must not become an excuse for the rest of the company not to understand and use artificial intelligence. Companies that quickly decide to hire such a person during this still quite enthusiastic period regarding AI must back that decision with concrete results, which are reflected in educated employees, client support, and expertise in the field. Will every director have to be an AI director in the future? Currently, we witness that most companies operate successfully without them, but given the accelerated development of artificial intelligence, it is a question of which companies will be able to do so – and for how long.