Theoretical and Practical Principles for Generative AI in Communities of Practice and Social Learning

This chapter explores the implications of the integration of large language model AIs, such as ChatGPT, into communities of practice and social learning spaces, providing some preliminary suggestions for social learning leaders. Social learning is about grappling with uncertainty together, forming meaningful experiences through identity and activity, not just information transmission. AIs lack human-like vulnerability, emotions, and social engagement. They are reifications of the data on which they were trained and their creators’ intentions. Because they do not possess self-authorship, they are incapable of participation. Mistaking their reifications for genuine participation, this may distort the collective process of experiencing meaningfulness. Large language model AIs can be valuable tools in social learning, offering rapid access and manipulation of reified information for tasks like search, brainstorming, and summarization. However, social learning leaders should be vigilant in recognizing AI contributions and establishing critical distance by differentiating AI input from human contributions. Social learning with AI requires collective consent, transparency, and critical reflection on AI-generated responses, which can be achieved through an iterative, two-loop process of norming and use.