AI in QA — The Landscape — quiz

A QA lead pitches replacing the entire manual testing team with an LLM-based agent that 'tests like a human.' Which response is most grounded in the realistic capabilities of current AI?

• Agree — modern LLMs match senior manual testers across all exploratory and accessibility flows
• AI accelerates specific tasks (test generation, triage, locator healing) but does not yet reliably replace human judgement on novel exploratory testing, UX, or domain-specific risk analysis
• Reject — AI has no place in QA workflows because it is non-deterministic
• Replace only the automation engineers, since AI cannot help with manual work