Turn on AI (or how to time implementation)

The authors of this article, Filip Tichý (Partner at Grant Thornton Slovakia) and Jakub Chudík (Co-Founder at Assetario), take you through the world of artificial intelligence in the AI Breakfast series. This article was written without the use of AI.

The history of innovation shows that being the first to market can bring significant success and market dominance, but also great risks and even existential threat. Typical examples of these extremes include Amazon, which transformed retail into e-commerce in the 1990s, becoming a technological hegemon for decades to come. On the other hand, there were the first social networks, which no one knows about today (e.g., MySpace), that did not turn into "Facebook." Autonomous or flying cars already exist technically, but many companies will go bankrupt in their efforts to bring them to market before these promising products are successfully commercialized. Simply put, being first does not guarantee success.

What about the implementation of artificial intelligence into business models in various industries? Will the early movers gain a decisive competitive advantage in their sector? Or will early adoption not change much, and it will suffice to wait until AI technology is mature and accessible and then simply turn on AI "plug and play" style? In the case of other general purpose technologies such as electricity or the internet, early adoption had of course helped many companies, but no insurmountable barriers have been created for "late adopters." These are now freely available commodities that can be used anytime by anyone. That is why there are two different approaches to AI: trying to implement AI as soon as possible to gain a decisive competitive advantage, or not wasting resources and waiting until the market and customers for AI in my industry mature.

Adopting AI does not just mean introducing new software, but also requires a fundamental change in the business model, organizational structure, and corporate culture. Companies that successfully integrate AI often transition to more agile, data-driven processes and emphasize continuous learning and adaptation. Companies should therefore view AI not as a one-time investment, but as a long-term transformation process.

The key to success: timing and convergence of innovation

When making strategic decisions and timing the implementation of new technology (such as AI), it is important to monitor innovation across the entire industry ecosystem, not just a single area. Typically, a breakthrough in an industry occurs when there is a convergence of several transformative innovations within that industry. Take autonomous vehicles, for example: development in this area has been ongoing for several decades. The first players went bankrupt long time ago. Most of the hardware is in place, as is instant and fast communication via 5G or GPS. However, other elements still need to mature (in terms of technological sophistication and economic accessibility) – AI models capable of autonomous "driving," special sensors, infrastructure, even greater quantities of data, and more. Each of these elements has its own trajectory of development and affordability in terms of cost.

The key factor for the successful adoption of AI in our industry is proper timing—not just being among the first, but also being ready for the moment when technology, the market, and the company's internal capabilities meet at the right point. Historical experience shows that it is often the "convergence of innovations" – the interconnection of several innovations relevant to my industry – that is decisive. It is also necessary to seek a competitive advantage in the unique use of AI and proprietary data, not just in the speed of adoption.