An online educational platform is using a foundation model to power its AI tutor, which helps students by answering questions related to math, science, and history. The company wants to ensure the AI behaves responsibly—specifically, that it does not generate offensive, biased, or factually incorrect responses, which could harm student trust and violate ethical guidelines. What should the team implement to ensure safe and responsible content generation?