Product iteration speed is the compounding variable in consumer software. The elapsed time between a product decision and a live deployment accumulates in code review queues and testing backlogs, and raw engineering headcount rarely solves either. Grab's chief financial officer said Tuesday that AI has cut that cycle by more than 30% at the company, a figure disclosed alongside second-quarter results and a raised full-year forecast.
Where the time comes from
In software delivery, the gap between a decision and a deployed product is rarely what it appears on a project schedule. A feature can take days to build while spending weeks waiting on reviewers or queued for QA before release. The total cycle time, measured from scoping to live, typically has engineering work as its minority component. AI tooling applied across the overhead stages can pull calendar time out of the cycle at friction points where more engineers do not help. Grab's CFO credited AI with the more than 30% gain but did not, in the available remarks, specify which stages of the development pipeline account for the largest reductions.
The forecast revision
Grab reported second-quarter results on Tuesday and raised its full-year outlook alongside the product-speed disclosure. The CFO placed the AI efficiency improvement directly inside the commercial argument, tying faster shipping to the revised guidance. The company did not break out where the 30%-plus acceleration is concentrated across its product lines.