294 patients with neck and/or low back pain were randomized to receive standard management alone (n=97) or in addition to an AI-based, individually tailored self-management application (n=99) or an e-Help web-based support tool (n=98). The primary outcome of interest was musculoskeletal health. Secondary outcomes of interest included disability, pain outcomes, perception of illness, quality of life, overall improvement, and adverse events. Outcomes were assessed at 6 weeks, 3 months and 6 months. The AI-based application provided no significant advantage over the web-based support group or the usual care group with respect to musculoskeletal health. Overall, very little differences were observed between the two groups, suggesting that an AI-based self-management applications, as developed for this study, may not provide advantages over current management pathways for neck and/or low back pain.