Participatory planning theory evolved across several decades through a wide body of often fragmented literature. Foundational traditions, such as Davidoff’s advocacy planning, Arnstein’s ladder of citizen participation, Forester’s communicative planning, Healey’s collaborative planning, and Innes and Booher’s collaborative rationality, provide distinct though often competing interpretations of what meaningful participation consists of and the conditions that are necessary to attain it. This fragmentation produces several structural challenges as Generative AI is increasingly grounded within participatory planning practice. In this context, the discipline lacks an integrated basis for evaluating its implications. Existing research on Generative AI focuses predominantly on technical capabilities, providing limited engagement with the normative positions of participatory planning, as well as insufficient attention to questions of legitimacy and power.

This research argues that the participatory traditions, taken collectively, contain the resources required to assess Generative AI, but only if they are synthesised rather than applied individually. Applying Jabareen's (2009) conceptual framework methodology, the paper derives five evaluative dimensions from undertaking a systematic comparison of the traditions: (1) Access and Representation, (2) Influence and Power, (3) Communication and Trust, (4) Deliberation and Inclusion, and (5) Legitimacy and Outcomes. Each dimension emerges from a pattern of convergence and divergence across the five traditions and generates a distinct set of evaluative questions for Generative AI tools in participatory planning. The framework is presented as a conceptual contribution in its own right, with empirical research identified as the next step to test and validate its dimensions.