Steve Watson emphasizes the vital role of critical thinking for testers in the AI era, advocating collaboration with AI tools rather than replacement. The discussion highlights challenges like AI biases, the need for human oversight, and adapting testing roles to focus on validation, analysis, and training future generations with balanced AI and critical thinking skills.

Software Testing Unleashed

Software Testing Unleashed - hosted by Richard Seidl. Different guest per episode. The official Show notes contain a comprehensive overview of the episode. Released as audio and video.

• https://www.richard-seidl.com/en/testing-unleashed
• https://www.youtube.com/playlist?list=PL48Mbm-L0hjB1OdwYi9h7jrq9t352-Zk_

Episode Details

• Show Notes: https://www.richard-seidl.com/en/blog/ai-critical-thinking
• Published: 2026-02-05T05:00:00Z
• Duration: 1661
• Author: Richard Seidl | Software Development & Testing Expert

Overview

The podcast explores how critical thinking remains essential in software testing as AI tools become more involved in the process. Rather than replacing human testers, AI is presented as a tool that enhances their work, shifting the role from giving direct instructions to collaborating with AI systems. This transition brings challenges, such as the potential for AI to inherit biases from its training data and the necessity for testers to verify AI-generated outputs. Maintaining strong analytical and questioning skills is emphasized, as testers must treat AI as an assistant rather than a definitive source of information.

The discussion includes examples of AI’s capabilities, such as efficiently analyzing survey data and generating test code, but also highlights limitations like missing ambiguities and the need for human oversight. As testing responsibilities evolve, the focus is shifting toward validation, understanding user needs, and adapting training to prioritize critical thinking and methodology over technical coding skills. The future of testing is viewed positively, with an emphasis on continuous learning and collaboration between experienced professionals and those entering the field.

What If

• What if you used AI to generate test scenarios but added a mandatory manual review step for critical edge cases?

  • Move: Integrate AI for drafting test scripts (e.g., Selenium) but allocate time for a weekly deep-dive session to manually validate AI-generated test cases for bias or missed edge cases.
  • Why now: The text highlights risks of AI biases and missing ambiguities, as seen in AIs analysis of survey data. Manual review ensures your testing doesnt overlook demographic gaps or nuanced requirements.
  • Expected upside: Higher confidence in test coverage, reduced risk of flawed AI outputs, and a competitive edge in delivering quality despite rapid AI adoption.

• What if you created a personal critical thinking framework for evaluating AI outputs?

  • Move: Develop a checklist (e.g., Does this AI output question assumptions? Are references provided? Are edge cases considered?) to assess AI-generated code or analysis before deployment.
  • Why now: The text stresses that testers must question AI like they do requirements. This ensures you dont rely on AI as a definitive truth without scrutiny.
  • Expected upside: Enhanced ability to spot flaws in AI-generated outputs, stronger validation practices, and alignment with the fields shift toward human-analytical depth over automation-only.

• What if you implemented a hybrid training plan for your skills that integrates AI tools and critical thinking workshops?

  • Move: Dedicate 20% of your weekly learning time to mastering AI testing tools (e.g., Copilot, AI test generators) and 20% to workshops on ethics, bias detection, and requirement questioning.
  • Why now: The text emphasizes the need to train the next generation with both AI and critical thinking. As AI evolves, solo developers must stay ahead by balancing tool use with foundational skills.
  • Expected upside: Future-proof your expertise, adapt to shifting roles in testing (from coding to validation), and position yourself as a thought leader in the AI-integrated software quality space.

Takeaway

• Repurpose critical thinking skills for AI collaboration
  Shift from instructing AI tools to actively questioning their outputs, assumptions, and data sources. Verify AI-generated test code, summaries, or insights by cross-referencing with manual analysis to ensure accuracy and avoid biases in training data.

• Validate AI-generated test code with manual oversight
  Use AI to automate code generation (e.g., Selenium scripts), but manually review and test the output to ensure quality, catch ambiguities, and confirm alignment with requirements, as AI may overlook critical edge cases or biases.

• Focus on high-level validation over manual coding
  Leverage AI for repetitive tasks like code writing or data analysis, freeing time to focus on validating end-user outcomes, questioning AI assumptions, and ensuring the product meets real-world needs beyond technical correctness.

• Prioritize training in critical thinking and AI literacy
  Invest in learning how to use AI tools effectively while maintaining analytical skills. Emphasize teaching new hires (and yourself) to question AI outputs, understand biases, and apply testing principles to AI-generated solutions.

• Use AI for efficiency but retain human-in-the-loop validation
  Adopt AI to analyze data (e.g., survey results) for patterns, but manually review AI-identified themes for context, missing nuances, or misinterpretations. Always maintain a balance between AI assistance and human oversight in testing workflows.