As generative artificial intelligence (AI) becomes increasingly integrated into various industries, media coverage plays a crucial role in shaping public perceptions of this emerging technology. This study examines how partisan media outlets frame generative AI in news headlines and social media descriptions and how these frames influence user engagement and emotional reactions. Using text analysis and Latent Dirichlet Allocation (LDA) topic modeling, we found that while both headlines and social media descriptions covered similar generative AI topics, social media descriptions featured a broader range of keywords but omitted discussions on AI’s social benefits, suggesting a shift toward engagement-driven content. Additionally, conservative media framed AI with more positive and trusting language than liberal media, challenging assumptions that conservatives are generally more skeptical of emerging technologies. Liberal media audiences engaged more positively (likes, shares, “love,” and “wow” reactions), while conservative audiences expressed more anger. These findings contribute to framing theory by demonstrating how partisan leanings shape AI discourse and public engagement across platforms. The study highlights the need for balanced AI reporting and greater awareness of how ideological biases influence media narratives.