The Information Freak
Sergey's management philosophy evolved through different contexts but centers on one core principle: transparency over control:"I always explain to teams that I'm an information freak, not a control freak."For Sergey, running a business is like driving a machine — you need to know exactly which levers to pull and what happens when you pull them. This approach requires clear responsibilities and reliable execution:
For me, transparency is extremely important. When a team member says something, they do it — you give a signal and know what response you'll get, not guess.
The Favorite Yango Value
Central to Sergey's vision is Yango's core value of "go global, go local" — a principle that drives his team's development efforts. Unlike the tech industry's tendency toward one-size-fits-all solutions, he believes in the power of cultural adaptation."In everything we do, it's extremely important when we take a global technology and adapt it to local needs. We often underestimate the specifics and small details in each country, each region, each city."One example of Yango's approach is Yasmina, a bilingual AI assistant powered by YangoAI, extensively trained on Khaleeji content and refined through input from Arabic-speaking experts.*YangoAI is powering technology that is fine-tuned and continuously trained on region-specific data. This allows Yango Group’s services to better reflect local culture and user expectations across the Middle East.
If you missed our article about Yasmina with Swithin D'Silva, Head of Smart Devices for MENA at Yasmina, catch up here.
Most solutions are rather one-size-fits-all, but not all people should eat the same food. The whole world doesn't eat Italian or Chinese cuisine — everyone finds something of their own.
Beyond Foundation Models
The AI landscape has been dominated by foundation models — massive neural networks trained on enormous datasets capable of generating content, translating languages, writing code, and creating music and images. According to the AI Index Report 2025, model scale continues growing rapidly, with training compute doubling every five months.But Sergey sees this era reaching a plateau: "Some say AI development has reached its peak, and smaller, more specialized models are required." The challenges are real — lack of available data, computing power constraints, and energy consumption that makes a single ChatGPT query consume about 10 times as much energy as a Google search.Building foundation models requires massive infrastructure investment. Epoch AI estimates that training frontier AI model costs have grown two to three times per year for eight years, with the largest models expected to cost over a billion dollars by 2027.
Robotics in Action
Here's where things get really interesting. Yango's autonomous delivery robots in Dubai exemplify this vision — AI-powered systems that navigate complex urban environments, delivering within two kilometers in under 30 minutes while supporting sustainability goals.
The robots offer contactless deliveries while helping reduce emissions and urban congestion. It's AI that directly improves people's daily lives.
We're transitioning from an era of competing to build the largest foundational models to one focused on delivering practical, intelligent AI agents that can reason, plan, and act autonomously.
The Burning Question
On the question 'Will AI replace us?', Sergey frames AI's impact on human roles as augmentation rather than replacement:
I don't look at AI as something that will replace humans. It's a new tool that will allow people to create completely different things. The most important thing AI won't replace is creativity — the ability to figure out what questions to ask and what target picture you want to achieve.
To keep our jobs, we should stay creative and sensitive: "AI plays a role in ad moderation and fraud detection. At the same time, humans still excel at understanding local contexts, cultural nuances, and emotional aspects of marketing."He sees professions evolving rather than disappearing: "What is a profession? It's performing a specific function. Some functions will become much easier to perform automatically with AI. But humans as a species will be able to create new things using these tools."
Advice for Businesses in the AI Era
Looking ahead, Sergey sees businesses at the beginning of a major shift. His advice is both practical and philosophical:"The most important thing for businesses is to step back and constantly ask: is there something we can do differently using current technologies? You need to be flexible and not afraid to change established principles."He warns against AI implementation for its own sake: "Sometimes it goes to what I consider overheating, when businesses force teams to use AI where it's needed and not needed. This is rather a result of FOMO — Fear of Missing Out."
Staying Fresh in High-Pressure Roles
Sergey's approach to avoiding burnout combines daily essentials with complete periodic disconnection. His foundation includes sports, sleep, and daily pleasures like good food. But the crucial element is genuine vacation time."You need to give your brain time to organize information, and this often leads to coming back much more productive," he describes his approach of physically removing devices and checking them only a few times daily during breaks.
The Yango Way Forward
For Sergey, being part of Yango means embodying the company's values while creating meaningful impact:
We make life better, simpler, and more pleasant. When people say that in Côte d'Ivoire, Yango has become a household name, it means we've truly made many people's lives better.
His vision of locally-relevant, globally-capable AI represents a thoughtful approach that could reshape how the industry thinks about cultural adaptation and practical application. Through his leadership at Yango, Sergey continues building AI solutions that serve communities rather than forcing communities to adapt to technology.As AI continues evolving from impressive demonstrations to practical daily tools, leaders like Sergey Loiter are proving that the future belongs to those who can combine global technological capabilities with deep local understanding.
