Added value instead of hype: how human-centered AI succeeds

Between hype and reality

AI Applications let us marvel: With more impressive speed, AI answers questions, selects suitable content, supports decisions, helps with complex tasks, and completes them faster. It saves costs, supplies data and works innovative. But: These advantages arise not automatically. They only arise when the technology is used sensibly – from the user’s perspective.

What users really need

AI applications make us doubt: when they give strange answers, when they “hallucinate,” when they don’t understand our concerns, and when they confuse us more than they help us.  
That’s why at UID AI, we think not only technically, but above all humanely. We design onboarding, language, feedback, and transitions in such a way that users feel understood. And we ask critically: Is AI really needed here, or perhaps something else entirely? User research is key to market- and user-oriented solutions.

AI with UX-DNA

Our experience shows that successful AI projects don’t start with a tool, but with a question: What problem do we want to solve, and for whom? Only once that is clear do we work with our customers to develop solutions that really help—and don’t just look modern. 

 
In doing so, we pay attention to:

• Transparency: What can AI do—and what can’t it do?
• Accessibility: Simple language, barrier-free design, clear user guidance 
• Trust: Feedback options, source references, handover to real people
• Embedding: How can AI be meaningfully integrated into existing processes? Service Design is unbeatable in this context. 

From practice: Examples of customer projects using AI

Orientation in production

In a production environment, AI helps to reassign lost components to the correct order. The employee photographs the part, the AI visually compares it with current orders — including size comparison — and suggests an assignment. This is not a classic chatbot, but rather an intelligent assistant that works in the background and provides real relief.

Error analysis with recommendation for action

In another project, AI supports the analysis of technical errors. Users describe the condition of a device, and the AI provides possible causes and specific options for action—including a direct link to the appropriate function. The system greatly simplifies the search for information and goes far beyond traditional chatbots.

Predictive maintenance – with patience

This example shows how important the implementation process is: AI is supposed to predict future wear and tear based on device data and suggest replacement parts in good time. However, AI is still in its infancy and lacks historical data. The added value for users is (still) low – and yet their data is needed to improve the system. Trust is crucial here: AI can only be successful in the long term if users understand why they should participate.

Q&A: Human-centered AI applications

Why is human-centricity so important in AI?
Because AI only creates real added value if it is geared towards the needs and expectations of users. Technology alone is not enough – the decisive factor is how it is experienced. Holistic and human-centered experience design provides decisive impetus for modern application development with AI.

What happens if AI is not designed with the user in mind?
Then it can confuse, frustrate or even be avoided – for example, if it gives the wrong answers, communicates incomprehensibly or offers no recognizable benefit.

How does UID ensure human-centered AI?
We design AI applications with a focus on access, language, feedback and transitions. We rely on user research to understand real needs – and on service design to embed AI into processes in a meaningful way.

Does every digital solution have to include AI?
No. We critically examine whether AI is really necessary – or whether another solution would be more suitable. The use of AI is not an end in itself.

What principles make AI applications successful?

• Transparency: Users understand what the AI can – and cannot – do.
• Accessibility: Clear language, barrier-free design, intuitive operation.
• Trust: Feedback options, source references, handover to people.
• Embedding: AI is a useful addition to existing processes.

How do you get started with AI projects?
Not with a tool, but with a question: What problem do we want to solve for whom? Only then do solutions emerge that really help. Still unbeatable: user-centered design with the phases of understanding, defining, designing and evaluating.

Curious? 

Thinking about how AI can be put to good use in your company? 
Or whether a chatbot, an assistant or another form of AI is really the right solution? 
 
Let’s talk. Together, we’ll find out what your users really need – and how AI can help.  

Lisa Reimer has been a Senior User Experience Consultant for over 15 years, supporting clients from various industries on their journey from the idea to the finished product or service. She primarily designs and evaluates suitable user interfaces. She also enables project teams to work innovatively and agilely. She uses the co-creative process LEGO® SERIOUS PLAY®, for example, to promote new processes and ideas and to make collaboration inspiring. As a speaker at various events, Lisa passes on her knowledge of environment design and digital transformation.