AI ethics, in data science terms, refers to the principles and guidelines that govern the development and deployment of artificial intelligence models to ensure fairness, transparency, and accountability. It underscores the importance of creating algorithms that don't perpetuate societal biases or discriminate against specific groups. Ethical AI demands transparent machine learning processes, where the decision-making of models is understandable and interpretable, not just a 'black box'. It emphasizes the necessity of informed consent, ensuring that data used to train models is obtained ethically, respecting individual rights. AI ethics also tackles issues of privacy, ensuring that personal data remains secure and is used responsibly. Furthermore, it promotes accountability, ensuring that when AI systems err or produce unintended outcomes, there's a clear line of responsibility. Ultimately, in data science terms, AI ethics is about harnessing the power of AI responsibly, ensuring that innovations benefit humanity without compromising individual rights or perpetuating harm.