Artificial intelligence will no longer change marketing — it already has. From video generation to automated campaigns, it’s accelerating both the pace and the way brands communicate and sell. The talk showed practical examples of why that’s good news for some and worse news for others.
AI in marketing: from hype to practice
A spring survey says that roughly 44% of Slovak marketers use AI daily, with smaller companies deploying it more boldly than corporates. Large organizations are slowed by security and internal rules, while small ones seek speed and a competitive edge. The momentum has advanced all the way to video, where AI comes up with the script, dialogues, and lipsync and infers context — for example, mud on the 'camera' in an off-road shot or sound effects, even though no camera exists. In practice, that means faster production of content that feels authentic.
Alongside ChatGPT there are multiple language models for text, data, and analytics that cover most tasks well enough. Europe, however, is lagging: key development and capital are in the USA and China, which means weaker domestic models and sensitive data traveling halfway around the world. Companies therefore often ask at least for access and APIs to foreign systems. The reality is unpleasant, but it forces rapid adoption of the tools that are available.
A new paradigm for search and shopping
AI overviews in search move the answer before the click and change the rules of the game for SEO, content, and online stores. The user wants a solution immediately and without extra steps, which is a hard blow for websites built over many years. Adapting means thinking in answers, not links, and optimizing for AI conversations as well. At the same time, new business opportunities are opening directly in chat interfaces.
Integrations like ChatGPT + Shopify show a purchase in a few seconds: recommendation, selection, confirmation — without a cart and without clicking through, because AI knows preferences and the platform has the data. AI agents are arriving that will handle a phone call to a service center or hair salon and book an appointment with a single approval. Amazon is testing 'Buy for me': if it doesn’t have the product, it will find it elsewhere, sell it, and take a commission in order to remain the customer’s first choice. The web as a network of pages is thus giving way to services where agents do the work for us.
Images, campaigns, and added value
In images, a big change came thanks to a system from Google nicknamed Nano Banana, which maintains product consistency across different scenes. From the same bar you can get realistic mountain photos or creative variations, and MidJourney can go from photorealism to fantasy. AI can also handle colorizing old black-and-white shots or creating visuals for 'juices full of fruit'. All of this now works for about 20 € per month without special hardware.
In a campaign for Nivy, AI automatically replied to contest participants and asked for a follow, which brought 31 % more followers; setup took three hours, the rest ran on its own. With Lira’s Advent calendar, the best-performing asset was a properly shot photograph, but second place went to the AI variant — a signal that the combination of approaches is strong. The point of the talk is captured by the sentence 'AI took my job to the next level': it’s not about replacement, but about amplifying performance, speed, and creativity. Those who learn to use the tools will move their work and their clients’ results a tier higher.
