Hey everyone,
AI is becoming a big part of software development…, and as testers, we need to adapt. Traditional testing methods don’t always apply when dealing with machine learning models, dynamic outputs and evolving AI behaviors.
I have been working on testing an AI-powered chatbot…, and I have faced challenges like:
- Non-deterministic responses – The bot doesn’t always return the same output for the same input.
- Bias and fairness testing – Ensuring the AI doesn’t unintentionally discriminate.
- Performance under load – Handling unpredictable AI response times.
I would love to hear from testers who have worked on AI applications. What strategies, tools, or frameworks have you found useful: ?? How do you handle edge cases and maintain test coverage for an ever-learning system: ?? I have already gone through this thread https://club.ministryoftesting.com/t/day-13-develop-a-testing-approach-and-become-an-ai-in-testing-champion-react-native-course but still looking for some more help.
Looking forward to your insights !!
With Regards,
Derek Theler