While the rise of artificial intelligence (AI) tools holds promise for delivering benefits, it is important to acknowledge the associated risks of their deployment. In this article, we conduct a focused literature review to address two central research inquiries concerning ChatGPT and similar AI tools. Firstly, we examine the potential pitfalls linked with the development and implementation of ChatGPT across the individual, organizational, and societal levels. Secondly, we explore the role of a multi-stakeholder responsible research and innovation framework in guiding chatbots’ sustainable development and utilization. Drawing inspiration from responsible research and innovation and stakeholder theory principles, we underscore the necessity of comprehensive ethical guidelines to navigate the design, inception, and utilization of emerging AI innovations. The findings of the focused review shed light on the potential perils of ChatGPT implementation across various societal levels, including issues such as devaluation of relationships, unemployment, privacy concerns, bias, misinformation, and digital inequities. Furthermore, the proposed multi-stakeholder Responsible Research and Innovation framework can empower AI stakeholders to proactively anticipate and deliberate upon AI’s ethical, social, and environmental implications, thus substantially contributing to the pursuit of responsible AI implementation.