A senior tester’s take on why AI isn’t replacing us—it’s making us superheroes
As someone who’s been elbow-deep in bug reports and test cases for more years than I care to count, I’ve watched plenty of “revolutionary” tools come and go. Remember when record-and-playback was going to solve all our problems? Yeah, we’re still waiting for that magic bullet. But here’s the thing about AI in testing—it’s not trying to be our replacement. It’s becoming our external imagination.
The Real AI Revolution Isn’t What You Think
Forget everything you’ve heard about AI automating testers out of jobs. Maaret Pyhäjärvi from CGI has been experimenting with AI in testing since she landed a 3.6 million euro research grant six years ago, and her insights are eye-opening. The real power of AI isn’t in cranking out more automated scripts—it’s in thinking differently about testing itself.
Picture this: You’re staring at a new application interface, and you’ve been looking at it so long that you can’t see the forest for the trees. That’s where AI shines. Pyhäjärvi describes showing ChatGPT screenshots of applications and simply asking, “What do you notice here? What looks wrong?” The results? Sometimes better than what human colleagues catch. It’s like having a fresh pair of eyes that never gets tired, never gets tunnel vision, and always brings curiosity to the table.
From Code Generation to Intent Exploration
Here’s where most people get AI wrong in testing. Sure, tools like GitHub Copilot can help you write test code faster, but that’s missing the bigger picture. As Pyhäjärvi puts it: “Code always follows intent. So intent is more important to me”.
The magic isn’t in generating more test scripts—it’s in using AI to explore intent, uncover edge cases, and spot gaps in our thinking. When you’re doing exploratory testing (you know, that beautifully chaotic dance of curiosity and systematic investigation), AI becomes your thinking partner, not your replacement.
The Practical Side: Starting Small, Thinking Big
The best advice from the trenches? Start where you are. Don’t wait for some grand AI strategy to be handed down from management. Instead, ask yourself: “Where do I lose four minutes every day?”
Maybe it’s writing up those repetitive exploratory session notes. Maybe it’s updating object maps when the UI changes. Maybe it’s generating test data that actually makes sense. Start there. Use AI to handle the tedious bits so you can focus on what humans do best: creative test design, risk analysis, and asking the tough questions that uncover the really interesting bugs.
At CGI, they’ve built their own GenAI assistant called CGI Navi that works with local LLMs through tools like Ollama, keeping proprietary data secure while letting testers experiment. They’re using it for GUI testing workflows, requirement analysis, regression test reviews, and test data generation. Nothing fancy—just practical applications that make daily testing life better.
The Reality Check: It’s Not Perfect (Yet)
Let’s be honest here—AI testing tools aren’t magic. Pyhäjärvi mentions Testzeus Hercules, which can take Gherkin scenarios and figure out how to operate websites automatically. Sounds amazing, right? But she’s quick to add: “It’s not perfect. I don’t fully trust it yet, and it sometimes costs more than I’d like—but it’s exciting”.
This is the mature approach we need. AI is a powerful tool, but it’s still a tool. The tester’s judgment about what’s worth testing, what risks matter, and where to dig deeper—that’s irreplaceable human skill.
The Bottom Line: Raising the Ceiling, Not Lowering the Floor
Here’s the truth that should excite every tester reading this: AI isn’t replacing testers—it’s raising expectations of what skilled testers can do. The real value lies in how we use these tools to amplify our thinking, expand our coverage, and ask better questions.
The future belongs to testers who embrace AI as their external imagination—those who use it to see patterns they might miss, explore scenarios they wouldn’t think of, and free up mental bandwidth for the strategic, creative work that makes great testing great.
So don’t fear the AI revolution in testing. Embrace it. Start small, experiment freely, and remember: “AI won’t replace you, but someone using AI just might”. Time to make sure you’re in the second category.
Want to dive deeper? Check out the full panel discussion and related content on maximizing AI’s potential in quality assurance—because the learning never stops in this field.