This research presents a novel two-step approach for generating acceptance tests for web applications using Large Language Models, addressing a significant gap in automated testing.

• Generates natural language test scenarios from user stories
• Converts these scenarios into executable test scripts
• Achieves 78% usability in a real industrial context
• Reduces testing effort while maintaining quality

For engineering teams, this approach offers a practical solution to streamline quality assurance processes, enabling faster development cycles while maintaining robust testing coverage.

Acceptance Test Generation with Large Language Models: An Industrial Case Study