In May 2025, the EuroSTAR conference once again brought the European software testing community together - this time in Edinburgh under the motto "AI on Trial". With hundreds of experts, speakers and thought leaders, it was all about the growing role of artificial intelligence in testing. It was a critical, thoughtful and honest look at what AI means for our industry: both in the testing of AI systems and in the use of AI in testing practice.
For us at TestSolutions, it is essential to be informed about developments and trends in the industry. Innovation does not happen in a vacuum: only those who understand where testing is heading can actively shape this development - for themselves and for their customers. EuroSTAR 2025 offered numerous insights that will be directly incorporated into our work.
Florian Fieber represented us at EuroSTAR 2025. Below is his report.
Many presentations showed how AI is already influencing test processes today. Tools based on machine learning and LLMs (Large Language Models) generate test cases, analyze results and make intelligent recommendations.
But with these new possibilities come new challenges.
James Lyndsay's session "Test-driven Generation: A Hands-on Experience" offered practical experience on how to use GenAI for test generation in a meaningful way - including an examination of its limitations and peculiarities.
Similarly, in "RAG to Riches: Supercharging LLMs for Superior AI Testing", Venkatesh & Anamika Mukhopadhyay showed how Retrieval Augmented Generation (RAG) can significantly improve the quality of LLM responses in a testing context.
The more complex and commonplace AI systems become - for example in healthcare - the more important rigorous and ethically sound testing becomes.
Usha Kandala's presentation "The Hidden Code: Bias Testing for Ethical AI in Healthtech" highlighted how damaging unbalanced training data can be.
In his interactive session "Let's Test an LLM",Michael Bolton challenged participants to question the behavior of language models using exploratory testing.
In "Can We Measure Software Quality?",Bart Van Raemdonck posed the question of whether software quality can really be measured objectively - and reminded the audience that numbers alone are not enough.
Context, communication and judgment remain essential, especially with AI.
Despite the power of new AI tools, many presentations drew a clear conclusion: critical thinking, empathy and human judgment are and will remain the foundation of all quality assurance.
This was impressively demonstrated in the interactive debate "Team AI vs. Team Human" by Tuğba Karakaya & Viviane Hennecke: AI should be seen as a collaborator - not a replacement.
Rik Marselis & Wouter Ruigrok provided a strong counterbalance to the optimistic tones with "Five Reasons Why You Should NOT Use GenAI".
They addressed risks such as data protection, hallucinations and ethical gray areas - and underlined the importance of a reflective approach to AI.
Our aim at TestSolutions is to offer high-quality and future-oriented QA services. This includes evaluating and using new technologies responsibly.
EuroSTAR 2025 has confirmed several of our core beliefs:
Diversity in the testing profession was also a topic. Isabel Evans spoke in her keynote "Breaking Test Stereotypes: Who Is Testing and Why It Matters" about the diverse roles of testers and the power of diversity.
Abiola Rasaq showed the importance of inclusive leadership in QA in "Africa's Tech Growth: The Role of QA and Women Leaders".
Sessions such as "Feedback Loop from Hell" by Davy Kenis and "Surviving and Thriving with AI in QA" by Maryia Tuleika also took a practical look at organizational challenges, flawed processes and dealing with change.
Conferences like EuroSTAR are more than just sources of inspiration. They provide orientation, encourage reflection and have very practical benefits. They remind us why quality matters - and how our profession is evolving.
For TestSolutions, participation in such events is a targeted investment. It helps us to classify innovations, evaluate new tools in a meaningful way and actively shape the future of testing.
The message from EuroSTAR 2025 was clear: AI is on trial - and we testers are the ones asking the right questions.
Would you like to know how AI can be meaningfully integrated into your testing strategy?
Contact us!