This builds on request interception, so it is also something you run on a local project rather than in the editor here. The technique is standard on real teams, and interviewers ask about its trade-off often.
Your manager asks a reasonable question: "What does the app show when the server is down?"
You want to test it, but you cannot break the real server just to run a test.
Playwright lets you answer the request yourself. This is called mocking or stubbing: your test gives a fake answer instead of letting the request reach the real server.