AI-First Testing
summary
Claim: AI generates a fleet of ~40 virtual testers that, given only a URL, explore pages, write and execute test suites, find accessibility, usability and security issues, and summarize results. A 20-second ChatGPT query finds more problems than 50 humans in a workshop, and a visual-diff routine flags a Mother’s Day doodle as an expected change. The stack integrates LLMs with network and DOM data, enabling an AI-first testing loop where AI tests, reports JSON results, and feeds fixes back to AI developers.
AI-written summary · about
chapters (8)
- 0:00AI-First Testing intro
- 3:13Why AI-First is different
- 10:15Year of AI testing research
- 18:23AI-First Testing in practice
- 25:44AI impact on testing careers
- 37:28AI for exploratory & performance testing
- 42:03Future of testing with AI
- 49:54Adopting AI before devs
related talks
44mAI Empowered Software Testing
35mApp Testing with AI Tools – Make it Easy
32mQuality engineering in the AI age: From testing AI to leveraging GenAI
26mAI and the Future of Testing
29mQA vs AI: Friends or Foes?
43mAI for Testing – Panel Discussion
43mExploratory Testing with AI
40mShould We Let AI Take Over Test Automation Completely?