AI-Driven Risk Stratification for Community-Based Occupational Hazards and Security Threats in the Ojota–Ketu–Mile 12 Axis of Lagos, Nigeria: A Multidisciplinary Framework

African megacities demonstrate economic pulsing as well as increased exposure to occupational risks and security threats, with their urban transport-market corridors and manifolds. This paper used a multidisciplinary approach with a mixed-methods design of the Ojota-Ketu-Mile 12 axis of Lagos with the involvement of surveys (n = 283), key informant interviews (n = 30), and focus group discussions (n = 9), and advanced statistics and artificial intelligence (AI). In descriptive results of occupational hazards, musculoskeletal strain was reported as the most frequent hazard (44.9%), whereas respiratory stress was the second most frequent hazard (38.2%). In descriptive results of security, most frequently reported security hazards were theft (42.4%), followed by harassment (33.6%). Findings in the use of a chi-square and regression model were education, occupation, and gender as crucial predictors, and a three-factor exploratory factor analysis identified occupational strain, exposure to the environment, and vulnerability of security. The Random Forest stratifiers were ~86% accurate, and the AUC was 0.91, with stratification identifying 28.6% of respondents as high-risk and 41.3% as medium-risk, thus highlighting systemic vulnerability. These trends were supported by the qualitative evidence, which revealed how many people feel the strain, visibility, accidents, and insecurity, narrating how people feel cautious about adopting AI in society. The study concludes that mitigation must be an integrated intervention of environmental, ergonomics, and security interventions in combination with participatory and ethically-governed AI systems. These findings present practical directions to policymakers, civic administrators, and even market institutions that desire safer, healthier, and more resilient urban market spaces.