This study explores the pragmatic features of ChatGPT-generated responses, focusing on speech acts and politeness strategies. With the increasing use of AI language models in both communication and educational contexts, concerns have clearly appeared about their ability to produce language that is appropriate across different real-life situations. The study employs a qualitative descriptive design and examines 30 responses generated by ChatGPT across six communicative situations: requests, apologies, refusals, suggestions, expressions of gratitude, and complaints. The analysis was based on qualitative pragmatic analysis guided by Speech Act Theory and Politeness Theory as the main analytical frameworks. The findings show that ChatGPT produces familiar speech acts that are closely related to their communicative functions. Requests and suggestions are mainly expressed through directive acts, while apologies, gratitude, and complaints are primarily understood through expressive acts. Additionally, refusals are generally conveyed through commissive acts, often supported by expressions of regret or mitigation. Also, the responses commonly employ politeness strategies, with negative politeness commonly used in face-threatening or commanding situations, and positive politeness appearing in more supportive and cooperative contexts. Overall, the findings suggest that ChatGPT can generate pragmatically appropriate responses that generally align with well-known principles of speech act theory and politeness behavior, although its responses tend to follow predictable patterns and remain limited by contextual constraints................ Keywords:........... Pragmatics, ChatGPT, Speech Acts, Politeness Strategies, Applied Linguistics, AI-generated language.