GeoAI roadmap exposes colonial infrastructure biases in global transport, demands systemic equity reforms

Indigenous spatial knowledge systems, such as the Māori *whakapapa* mapping of land and waterways, offer non-extractive alternatives to GeoAI’s data colonialism. These systems prioritise relational ethics over efficiency, challenging the roadmap’s techno-solutionism. However, the roadmap’s focus on AI marginalises these epistemologies, treating them as 'legacy data' rather than living frameworks. The erasure of indigenous land tenure in transport planning perpetuates historical dispossession.

The roadmap’s equity focus echoes 19th-century railway imperialism, where European powers built transport networks to extract resources from colonies while sidelining local needs. The Bretton Woods system’s infrastructure loans in the 1950s-70s replicated this pattern, trapping Global South nations in debt for projects that served corporate interests. Today’s GeoAI risks repeating this history by optimising logistics for capital flows, not human dignity. The roadmap ignores how colonial transport legacies (e.g., apartheid-era highways in South Africa) still shape urban segregation.

African *Ubuntu* philosophy centres collective well-being in transport design, contrasting with GeoAI’s individualised navigation. In Southeast Asia, *gotong royong* (mutual aid) logistics networks demonstrate how communities share resources without corporate intermediaries. The roadmap’s Western-centric metrics (e.g., 'efficiency') fail to capture these systems’ resilience and equity. Cross-cultural comparisons reveal that 'equity' is not a universal concept but a negotiated outcome of local power dynamics.

GeoAI’s bias toward quantitative data overlooks qualitative metrics like community trust or ecological impact, which are critical for equity. Studies show that algorithmic transport models disproportionately disadvantage low-income and minority groups due to biased training data. The roadmap’s emphasis on 'equity-focused AI' lacks peer-reviewed frameworks for measuring structural inequity in infrastructure. Scientific literature on transport justice (e.g., Sheller 2018) highlights the need for participatory design, which the roadmap sidesteps.

Artistic interventions, such as the *Transit Equity Atlas* (2022), use cartography to visualise racial and class disparities in transit access, challenging GeoAI’s sterile data visualisations. Spiritual traditions like the Hindu concept of *dharma* (duty to society) critique the roadmap’s profit-driven transport ethics. Performance art in favelas (e.g., Rio de Janeiro’s *Teatro do Oprimido*) reimagines public space as a site of resistance against AI-driven gentrification. These approaches foreground embodied, lived experiences over abstract metrics.

Scenario modelling shows that GeoAI could deepen inequality if deployed without community control, as seen in predictive policing’s racial biases. Alternative futures include 'commons-based logistics' (e.g., Barcelona’s *Superblocks*) or *buen vivir* (living well) transport models from Latin America. The roadmap’s techno-utopianism ignores the risk of AI being co-opted by authoritarian regimes to surveil dissent. Future-proofing requires embedding decolonial ethics into GeoAI’s design from the outset.

The roadmap’s 'equity focus' centres tech developers and policymakers, not the communities most affected by transport inequity. Women, disabled people, and informal workers—who rely on public transit—are sidelined in the narrative. Indigenous leaders in the Amazon have warned that GeoAI could accelerate land grabs for 'smart highways,' yet their voices are absent. The framing treats marginalised groups as beneficiaries, not co-designers of solutions.