AI has quickly become part of everyday life in modern business systems – but behind every intelligent function lies something even more important: data. The right data, the right structure, and the right governance. We spoke with Bogdan Dobondi, a data-driven development specialist at Implema, about why so many companies get stuck on the path to AI – and what it takes to make AI work in business systems.
Hi there Bogdan Dobondi, AI is taking its place in more and more business systems – but what does that mean concretely for companies’ daily operations according to you?
AI in business systems means we can make faster and more accurate decisions based on real-time data. It affects everything from automated forecasts and intelligent resource planning to smart warning systems. For this to work, however, data must be standardized, accessible, and of high quality. AI is no better than the data it’s fed – which is why data governance and data quality become critical factors for success.
What’s the biggest mistake companies make today regarding their data?
Many companies underestimate the importance of standardization and structured data governance. They collect large amounts of data, but without a common model, clear ownership, or governance structure. The result is silos, inconsistent data with low trust – which slows down both AI initiatives and digitalization in general. Here you need to start at the right end: build on standards, ensure data quality, and simplify access to data.
What do you see as the biggest risks when connecting AI with business-critical information?
The biggest risk is lack of control. When AI agents are connected to business-critical information without clear governance, it can lead to incorrect decisions, security breaches, or unintentional data leakage. It’s crucial to have transparent processes, traceability in AI agents with clear boundaries for what data can be used, and how. Additionally, we must consider regulatory requirements, like GDPR, from the beginning.
How do we at Implema work to ensure that customer data is not misused in AI models or external integrations?
We consistently work with standardized processes for data management and governance. This includes defining clear roles, access control, and ensuring traceability and transparency in data flows. When we develop AI solutions or external integrations, security and integrity are always built in from the start – we follow the principle of “privacy by design.” Our goal is clear: the customer should always have full control over their data.
What’s the most important thing companies can do today regarding their data? What’s the most important thing to tackle?
The absolutely most important thing is to establish a standardized data foundation. That means defining master data, structures, data quality, and responsibility. Without that, it’s difficult to scale, automate, or benefit from AI. In parallel, you need to review data governance and security frameworks. By quickly gaining control over these areas, you create a stable platform for innovation and growth.
Finally, what are your future predictions regarding data, AI, and business systems?
We see a clear movement toward so-called “real-time decisioning,” where business systems and AI together enable fast, data-driven decisions based on real-time data (or near real-time). For this to work, data standardization, automation, and security will become even more important. At the same time, semantic data models and data architectures will enable both flexibility and control. The future belongs to companies that combine fast delivery with robust data governance.
