GrabAI

GrabAI is the artificial intelligence programme of Grab Holdings, the Southeast Asian superapp for ride-hailing, food delivery, payments and financial services. Machine learning systems developed under the programme underpin core Grab services including estimated time of arrival prediction, dynamic pricing, driver-rider matching, fraud detection and demand forecasting, and in the 2020s Grab expanded its AI efforts into generative AI assistants and merchant tools built on its in-house data platform.^[1][2]

History and Background

Grab was founded in June 2012 as MyTeksi, a taxi-booking application launched in Kuala Lumpur, Malaysia by Harvard Business School classmates Anthony Tan and Tan Hooi Ling, using prize money from the school's New Venture Competition. The service expanded regionally as GrabTaxi and was renamed Grab, and the company is now headquartered in Singapore with operations in Singapore, Malaysia, Cambodia, Indonesia, Myanmar, the Philippines, Thailand and Vietnam. Grab listed on the Nasdaq stock exchange in December 2021.^[1][2][7]

Grab has applied machine learning to its platform since its early growth phase, using it for travel time prediction, dynamic pricing, driver allocation, food delivery demand forecasting and fraud prevention. In the 2020s the company consolidated and expanded these efforts under the GrabAI umbrella, establishing an Artificial Intelligence Centre of Excellence (AI CoE) with support from Digital Industry Singapore, and in April 2025 it launched AI tools developed with OpenAI and Anthropic.^[2][3][7]

At its annual GrabX 2026 showcase in Jakarta, Grab unveiled 13 AI-powered features across consumer, travel, merchant and driver services, all powered by a proprietary Grab Intelligence Layer trained on insights from 20 billion rides and orders. The rollout built on an Early Access Program involving 200,000 users, with products released market by market subject to regulatory approvals.^[4][5]

Technology

Grab's AI stack combines traditional machine learning with large language models and computer vision. The Grab Intelligence Layer integrates real-world signals such as traffic, weather and foot traffic into automated decisions across the platform.^[5] Representative systems include:

• ETA prediction and dynamic pricing — models that estimate arrival times and adjust fares in real time based on supply and demand.^[2]
• Group Ride — a multi-agent routing system that sequences pickups for up to four passengers and calculates individual fares, promising up to 40% savings over solo rides.^[5]
• Grab AI Assistant — a conversational assistant that uses user habits and preferences to recommend food delivery, dining and shopping.^[4]
• Virtual Store Manager — a computer-vision tool that uses CCTV feeds to give merchants information about store traffic and operations.^[6]
• Driver AI Assistant — an assistant that answers driver-partner questions about trips and earnings.^[6]

Applications and Impact

GrabAI systems operate across Grab's eight markets, affecting everyday services used by millions of Southeast Asians: consumers receive faster and more reliable ride-hailing and delivery estimates, merchants gain AI tools for operations and marketing, and driver-partners receive guidance and income-support features. In travel, Grab announced AI-powered tools including GrabStays hotel booking, Discover by Grab restaurant recommendations and GrabPay for Travel cross-border payments. Grab's leadership has described AI as integral to the company's growth strategy, and the company has said it plans to introduce AI-powered delivery robots to speed up restaurant pickups.^[4][6]

See Also

• AI Automation
• Recommendation System
• Fraud Detection
• AI Startups Malaysia
• AI Malaysian Banking