Acceptance criteria and expected results define the benchmark for a successful test execution. Without them, there is no standard to compare the actual test output against, making it difficult or impossible to determine whether the test passed or failed. Other options are incorrect as tests can still be executed or automated without defined criteria.