As AI becomes part of how companies are discovered, compared and even judged, branding is taking on a new role. It is no longer only about how a company looks, but about how clearly and consistently it communicates across people, platforms and increasingly, machines. We spoke with Vedran Obradović, founder of Vizura Studio, about trust, B2B branding and what changes when AI becomes part of the audience.
How did Vizura Studio get started?
I have been working in branding and identity for more than eight years, first as a freelancer and later in lead and in-house roles, mostly for B2B and tech companies. I kept meeting the same situation: a strong product wrapped in a brand that explained it badly, and brand guidelines that lived in a PDF nobody opened after the first month.
Vizura Studio® has worked under that name since 2024 and is based in Zagreb. I run it alone on purpose, so the person who hears the problem is the person who designs the answer. The focus is B2B, tech and AI brands, and the work has grown from logos and guidelines into full brand systems, including the interactive Brand OS portals I build for clients.
What makes branding for B2B, tech and AI companies different from branding in other sectors?
The buyer cannot see what they are buying. A software, security or infrastructure product is invisible until someone trusts it enough to try it, so the brand does a lot of the explaining before any demo happens. There is also usually a committee, a technical person, a finance person and a founder, each looking for something different, and all of them comparing you with competitors who use the same gradients, the same abstract symbols and the same future-tech language.
The job is to be easy to understand and hard to confuse with anyone else. A distinctive look helps, but only after the message is clear.
One more difference is new. The brand now has two audiences, people and the AI tools those people ask. Buyers use ChatGPT, Claude and Perplexity to build shortlists, so the brand has to be readable for machines too.
What role does branding play in building trust in a company in these fields?
Trust forms before anyone can verify the deeper product claims. A buyer of a payment, security or AI product usually cannot audit the architecture on a first visit. What they can judge in seconds is whether the company looks and sounds consistent, mature and specific. A vague headline, assets that differ between the website, the deck and social, or an old logo in a partner’s slide all read as a signal about how the product itself is run.
So I treat branding as visible evidence of discipline: one clear positioning sentence, one visual system used the same way everywhere, and claims stated precisely. On Deepsea Networks, whose customers work offshore where safety matters, the website hero says “Operate safely. Be connected.” and every approved asset in the brand portal exists in one controlled version. Unapproved versions never leave the portal.
In your view, what do tech companies most often underestimate when presenting themselves to the market?
Consistency after launch. Most teams invest in the identity and lose it within months: sales builds its own slides, a partner gets an outdated logo, someone asks ChatGPT for a LinkedIn post and gets a generic voice. The brand drifts, and in a trust-heavy sector drift reads as carelessness.
The second is the plain explanation. Engineers know exactly what the product does and skip the sentence a buyer needs: what it is, who it is for and why it can be believed. That sentence also has to be on the website in a form AI tools can read, because that is increasingly where the first impression is formed.
Can you tell us about a project that particularly captures the way Vizura Studio works?
Deepsea Networks, offshore connectivity for Oil & Gas and Maritime. I built the logo (the idea is Signal + Waves = Handshake), the brand guidelines, a six-chapter 3D website and a live Brand OS. It shows how I work: the strategy decides the identity, and the identity is then built to be used, not just looked at.
The Brand OS is a browser portal for the team. Logo Studio gives every approved lockup for download as SVG or PNG. Pattern Lab generates patterns drawn from offshore data such as bathymetry, sonar and sea state, exportable at any size. Post Studio builds social posts from one set of copy and includes a check before publishing. An AI kit holds the brand context, so a team member using ChatGPT or Claude starts from the Deepsea voice and rules, and a Voice Check flags copy that drifts.
I built the same kind of portal for KORDIN, with its own rules, prompts and quality-check files for AI tools. Short walkthrough videos of both are available if useful for the article.
Artificial intelligence is changing both creative work and the way people discover and evaluate companies. How are branding strategies evolving in this context?
Two things are changing. Discovery: buyers ask AI tools for a shortlist before they visit a website. G2’s 2026 research reports that 51% of B2B software buyers now start research in an AI chatbot more often than in Google. Production: team members use AI to write posts, decks and images, and without the brand rules in the prompt the output sounds like everyone else.
This is why I build Brand OS, a brand guideline turned into an interactive portal with approved assets, templates and an AI context layer, so people and AI tools work from the same source. For the discovery side I built Brand MRI, a free scanner at labs.vizurastudio.hr. It checks a brand across seven layers (access, legibility, identity, clarity, evidence, signal and Brand OS) and shows what AI concludes about the company plus the three fixes that matter most. In my first test of well-known tech brands the best scored 56 out of 100 and my own studio scored 43. Most brands were built for human eyes only.
For fintech, where people are asked to trust a company with money or data, a practical first step is to run the scan, ask ChatGPT or Claude directly what they say about the company, and check whether the answer matches what the company wants to be believed.
