Product Launchclipcomputer visionurban safetyopen data
Researchers Deploy Safest Way App For Nighttime Safety
7.0
Researchers from UCL introduced Safest Way, a pedestrian navigation app using AI to suggest safer walking routes, launching a beta in early 2025 with about 1,000 users and notable uptake by Derry volunteers iCat. The team aggregated geospatial data—over 500,000 street lights, CCTV and police crime records—and used OpenAI's CLIP to model perceived safety from 500 photos, aiming to scale across the UK while noting prediction limits.
Key Points
- 1Deploys Safest Way app using AI to recommend safer pedestrian routes; beta reached about 1,000 users.
- 2Aggregates 500,000+ street lights, CCTV and crime data to compute safety scores for street segments.
- 3Enables practitioners to incorporate objective and AI-modelled perceived safety into route-planning and vulnerability training.
Scoring Rationale
Applied multimodal research with real-world deployment, but limited scale and single-source validation constrain broader impact.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
