Three perspectives on the user experience and AI: what UX + AI is in practice

Like any technology, AI-powered tools can either genuinely simplify daily work or frustrate users and slow down processes, which ultimately prevents them from gaining traction. But how to create AI tools with stellar user experience?

To illustrate how user experience and AI can come together in practice, I often use this simple model with our clients: UX for AI, UX by AI, and UX with AI.

Traditionally, user experience design in software development has focused on user interfaces and predetermining smooth user paths. The user directly controls the system, for example, through a Human-Machine Interface. Measuring success usually focuses on the software’s usability and how efficiently people use it.

AI systems change this dynamic: they operate based on probabilities, meaning their suggestions may vary even with the same input. This shifts the focus of user experience design to how the user interacts with the system and interprets results generated by the AI.

A good AI user experience enables transparency to the model’s behavior, making it explainable and easy to understand. A rational division of labor between human and AI is also a must: the AI assists, guides, and recommends – but leaves decision-making with the human. Measuring success is also different, as they focus on the accuracy and explainability of the AI-generated results – as well as the user trust and adoption rates that rise as a result.

Another angle examines improving the user experience of machines and systems by adding AI-driven functionalities. In OT/IT environments, adaptive dashboards that leverage real-time data, operator AI assistants, and trend-based prescriptive alerts can help machine operators at precisely the right moment. On the other hand, Edge AI, combining edge computing and artificial intelligence, can complement these solutions by running inference locally in the device itself. This minimizes latency and dependency on network connectivity, which are crucial for safety-critical and time-sensitive operations.

User experience improves when information comes at the right time and in the right format. The user can make decisions faster and avoid errors more easily. At the same time, their cognitive load is reduced. AI-driven functionalities do not only simplify individual tasks but also improve the overall production process. Success is tracked by following task completion time, number of errors, and user satisfaction. If these indicators improve, AI has made the work genuinely easier.

AI helps not just end users but also those designing digital solutions. Generative AI has proven particularly useful in ideation, analyzing large datasets, suggesting design alternatives, and improving documentation quality, for us at Etteplan as well. This also applies to other forms of design work: AI can improve specification quality by, for example, identifying deficiencies and contradictions. At the same time, it reduces errors and speeds up specification turnaround times. This allows designers and engineers to focus on higher-value tasks – the areas where humans excel.

The success of AI solutions is followed by tracking time to value (TTV), the amount of rework and user satisfaction. Exceling in these metrics leads to faster development cycles, reduced project risk, and improved customer experience.

When these three perspectives are considered together, AI solutions become both trustworthy and genuinely helpful in daily work. At the same time, design and development become more efficient, and solutions deliver business value faster. This isn’t just good design – it’s building competitiveness.

Which of these three perspectives does your organization need the most right now – building trust for AI, streamlining machine usage, or improving the efficiency of design and engineering work? Talk to our AI experts to explore how we can support your transformation.