BACKGROUND: The involvement of people with intellectual disability in research is framed as inclusive, denoting their active participation in its processes. However, questions are raised about ownership and control, genuineness of involvement, and the need for honest accounts to develop practice. Such issues are particularly pressing in Australia, where there is the absence of a strong self-advocacy movement to partner with academics or hold them to account. METHOD: Action research was used to reflect on and progressively refine the support provided by a research mentor to a co-researcher with intellectual disability employed on a large multimethod study. RESULTS AND CONCLUSIONS: Accepting the co-researcher's strengths and designing support on the job rather than teaching them to "pass" before venturing out in the field are important in ceding control. Support required for a co-researcher is more than practical and involves developing a relationship that can actively challenge views and foster reflection. Ownership of questions and disseminating of outcomes are hampered by contextual factors such as tender processes, short-term positions, and a failure to acknowledge the support required to present findings.