Grab's AI-Led Playbook Pays Off as Q2 Profit Jumps 620%
Source: Tech in Asia
Singapore's superapp has spent two years telling investors it treats AI as a margin lever, not a cost centre. Its second-quarter results, out this morning, are the strongest evidence yet that the bet is working — profit jumped more than sixfold, user numbers hit a record, and management raised its outlook for the...
Singapore's superapp has spent two years telling investors it treats AI as a margin lever, not a cost centre. Its second-quarter results, out this morning, are the strongest evidence yet that the bet is working — profit jumped more than sixfold, user numbers hit a record, and management raised its outlook for the rest of 2026.
Grab recorded a US$252 million profit for Q2 2026, up 620% from US$35 million a year earlier, on revenue of US$997 million — a 22% year-on-year gain spread across deliveries, mobility and financial services. Monthly transacting users reached a record 54 million. Chief executive Anthony Tan framed the growth as "led by transactions and users rather than prices — the healthiest form of growth we can generate."
The AI specifics are where the quarter gets interesting. Tan said the cost per AI interaction for drivers and merchants has more than halved since June 2025. Grab's engineers now pair with autonomous coding agents as standard practice, cutting time-to-market for new products by up to 30% year on year, while Jarvis — an internal AI data analytics assistant — saves sales teams roughly 40,000 hours every quarter. The company raised its 2026 revenue guidance to US$4.1–4.15 billion, lifted adjusted EBITDA guidance to US$720–740 million, and authorised another US$750 million in share buybacks.
The results land as Southeast Asia's platform economy absorbs a fuel crisis and currency headwinds, and as AI reshapes how regional tech giants operate. Grab's answer has been to embed AI across the stack — lending decisions in its digital banks, where total loans disbursed grew 72% to US$1.2 billion in the quarter, plus advertising and operational planning — rather than bolt it on as a headline feature.
Why it matters for Singapore: Grab is the closest thing the local ecosystem has to a homegrown AI-scale company, and its playbook matters beyond one profit-and-loss statement. If a Singapore-founded superapp can turn AI into measurable margin at this scale, it strengthens the case that AI adoption here can move past pilots into real commercial returns — exactly the gap government surveys keep flagging among the majority of local firms that have yet to adopt AI at all.
