Didi Chuxing held a safety open day to showcase its safety pre-warning system, which uses large multimodal models to scan for risks in millions of daily orders. The system prioritizes recall rate over precision, leading to many false positives but aiming to minimize missed risks. Didi invests heavily in this system, exceeding its net profit last year, and employs over 4,000 staff to handle risk orders, covering pre-trip, during-trip, and post-trip safety measures. AI
IMPACT Highlights the use of large multimodal models in safety-critical applications, demonstrating AI's role in risk management for ride-sharing services.
RANK_REASON Article details a specific product/system within a company, not a new release or major industry shift.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
