At the AI Weekend as part of the 17th Weekend Media Festival, a discussion titled ‘When do product companies bet on AI?‘ gathered two key speakers – Ivana Bešlić from Sofascore and Hrvoje Ćosić from Aircash. The founders of the two companies touched on the challenges and opportunities presented by artificial intelligence (AI), with a particular emphasis on the integration of AI tools into everyday business operations.
Bešlić pointed out that companies often fall into ‘hype‘ when it comes to AI technologies. Using his own experience in purchasing a washing machine as an example, he emphasized that AI is entering every aspect of our lives, but many companies lack the necessary experts to properly assess whether certain AI tools are useful or not. He specifically warned about the risks of using AI tools without a deep understanding of their functionalities.
‘If you don’t know how to drive a car and you buy a super-fast car, you will end up in trouble,’ says Bešlić. Similarly, without the appropriate knowledge about AI, companies may face challenges in managing tools that do not actually fulfill their promised functionalities.
– Companies do not have people who understand whether the tools are good or bad. When I use AI tools, I often get wrong answers, and if you do not have internal knowledge about what you are trying to improve, then you are in the fog and cannot assess whether the tool is useful or not, because you do not know the truth yourself – added Bešlić.
Ćosić built on this topic, emphasizing that pragmatism is a key phase in introducing AI. According to him, companies in the early stages of development may not have the resources to develop their own AI solutions, so they often rely on off-the-shelf tools. Aircash initially used purchased AI tools, but over time they developed internal capacities, including hiring engineers and experts for further development. Although AI is used, Ćosić emphasizes that in many cases, basic technologies still meet business needs. As for Aircash, AI is currently applied in the digital onboarding of users, but they are particularly interested in how AI can assist in monitoring and risk prevention, especially due to their collaboration with financial institutions.
– We need to be pragmatic about what ordinary technology can solve and what we actually need AI for – believes Ćosić.
‘I do not believe in the AI hype’
—
—
On the other hand, Sofascore also uses AI technology, but they do not rely on it entirely. For example, their chat where users can communicate during football matches is ‘protected’ by an algorithm that recognizes hate speech and all prohibited words on the platform. Additionally, the domestic company uses AI for player ratings in the app and in many other similar processes.
– I do not believe in the AI ‘hype’. Recently, I read an article by a person who tried to sum up all investments in AI and came to an interesting conclusion, which is that a huge amount of money is being invested in this technology, but many companies do not see returns. The question is whether as a company you should even buy AI tools or have internal knowledge, and realistically, you cannot do anything without knowledge. You need to have a team that knows what to do, what to ask, and what to expect. If everyone starts asking AI tools for help, and they are not critical of their answers, then that is not an ideal way of doing business – says Bešlić.
Both Ćosić and Bešlić highlighted the importance of education and the development of internal capacities within companies to maximize the potential of AI tools. Bešlić warned of the possible long-term consequences of over-reliance on AI tools without thorough understanding, predicting that ‘juniors’ who are learning with the help of AI may have limited skills when faced with more complex challenges in the future.
During the discussion, important questions were raised about the timing and manner of investing in AI technologies, especially for technology companies that find themselves at the crossroads between purchasing ready-made solutions and developing their own resources. The common conclusion was that, although AI is the future of business, companies must carefully consider how, when, and where to use it.
– Juniors will never become seniors because they are doing things with the help of AI from the start. It is possible that in 5 to 10 years there will be no developers who know their job well. Companies need to commit to employee education and invest in it as soon as possible – concludes Bešlić.
