פרק #68 | שילוב בינה מלאכותית בבדיקות אוטומציה עם גיל זילברפלד скачать в хорошем качестве

פרק #68 | שילוב בינה מלאכותית בבדיקות אוטומציה עם גיל זילברפלד

Integrating AI into Test Automation with Gil Zilberfeld In this episode, Gil Zilberfeld joins the show to discuss one of today’s hottest topics in software testing: integrating artificial intelligence into software testing and automation. Gil, a long-time quality expert, shares insights on how this technology is reshaping the tester’s role in a rapidly changing industry. Key topics covered: Should testers use artificial intelligence? The answer: it depends. Artificial intelligence can save time and help execute more tests in less time, but it still cannot be trusted completely. It should be treated as an assistive tool that supports - but does not replace - human judgment. How the tester’s role is changing: In the past, testers were mainly responsible for executing tests. Today, with artificial intelligence in the picture, testers are increasingly expected to manage and review testing activities performed or recommended by models. This requires a deep understanding of the system and business context - something an automated tool cannot fully provide on its own. Time optimization: Similar to automation, artificial intelligence is an investment that pays off over time. It enables testers to focus on more complex tasks while helping with test case creation, data generation, script writing, and more. Warnings and challenges: Artificial intelligence is not always accurate, and its suggestions can include errors. The tester must act as the "responsible adult" - filtering, judging, and validating the outputs. For example, if artificial intelligence generates tests that include dynamic fields (such as IDs or timestamps), testers must understand what can be compared reliably and what cannot. Common use cases: • Generating test cases based on API schemas • Refactoring test code using design patterns • Automatically producing documentation and defect reports (for example, a Jira ticket generated from a failed test) • Creating JSON files for required test data • Significantly reducing test development time (up to 70%-80% time savings) API testing vs. UI testing: API tests are generally easier to automate and manage due to clear schemas and contracts. UI tests are more complex and require more context, including understanding screen structure and user behavior. Using artificial intelligence effectively often requires an even stronger understanding of the system to "help it help us." Bottom line: Artificial intelligence is driving a major shift in the testing world, but it does not replace people. It strengthens our ability to test faster and deeper, while still requiring oversight, judgment, and a deep understanding of the product and processes. Link to our Community Whatsapp Group (https://bit.ly/TestIL_Whatsapp) LinkedIn profiles: • Gil Zilberfeld (  / gilzilberfeld  ) • Netanel Harush (  / netanel-harush  )