Grab builds in-house robotics to curb delivery costs

Increasing labor expenses and shrinking delivery profit margins are leading major platform operators such as Grab to explore automation. Their acquisition of robotic mobility startup Infermove represents a strategic step to build internal robotics expertise.

Grab operates at a scale where minor efficiency improvements can yield significant impacts. Its platform facilitates millions of deliveries across Southeast Asia, many completed by scooter and bicycle riders navigating dense urban environments. This complexity inherently limits how much automation can supplant human workers. By acquiring a company specializing in robots for unpredictable real-world settings, Grab signals its belief that physical-world artificial intelligence has matured sufficiently for deployment beyond controlled pilot programs.

Automating delivery near core business functions

Instead of depending on standardized, ready-made systems, Grab is choosing to internalize the development cycle. Infermove's technology is engineered to learn from actual movement data, including patterns from non-motorized delivery vehicles. Practically, this means robots are trained on how people genuinely navigate sidewalks, crosswalks, and crowded delivery points, rather than on idealized simulated environments.

For a delivery giant like Grab, this difference is critical. Simulations aid initial development but frequently fail to account for the unusual scenarios that define real-world urban logistics. Internalizing this learning process allows Grab to shape how automation functions within its specific operational limits, instead of restructuring its entire delivery network to accommodate a third-party solution.

From a corporate strategy viewpoint, the core value lies in enhanced control. Owning the underlying technology grants Grab greater influence over deployment speed, operational range, and cost-benefit decisions. It also diminishes long-term reliance on external vendors whose strategic goals may not align with Grab's regional presence or economic conditions.

Automation, however, is not framed as a wholesale replacement for human delivery personnel. Even as robots assume certain workflow segments, people remain essential to service fulfillment. Grab's focus seems aimed at targeted applications, such as repetitive, short-distance tasks in structured first-mile or last-mile segments. In these areas, robots could help manage demand surges, decrease delays during busy periods, and alleviate pressure during workforce shortages.

Controlling costs while maintaining service quality

During an internal meeting last December, Grab's Chief Technology Officer, Suthen Thomas, characterized Infermove's advancements as "impressive," noting both the technical capability and its early commercial application. He also stated the startup would continue operating independently, with its founder reporting directly to him. This structure indicates Grab prioritizes effective execution and business continuity over swift organizational merger.

This strategy mirrors a wider trend among large digital platforms. Rather than treating AI as a supplementary layer added to existing processes, companies are integrating it more deeply into fundamental operations. In delivery and logistics, this increasingly means advancing beyond pure software optimization into tangible automation, where risks and costs are greater but potential gains are more foundational.

The timing is significant. Demand for on-demand delivery keeps rising, yet profitability margins stay constrained. Customers anticipate quicker service and lower costs, while operators confront increasing wages, fuel prices, and regulatory scrutiny. In this climate, automation shifts from being a novel experiment to a necessary tool for maintaining service standards without sacrificing financial sustainability.

Locating robotics development nearer to core operations may also better align incentives around data utilization. Training physical AI systems demands vast quantities of real-world data, which delivery platforms like Grab already generate at massive scale. Maintaining this feedback loop internally can accelerate development cycles and minimize the need to share sensitive operational data with outside parties.

Limitations persist. Robots designed for sidewalks and short trips are not poised to completely replace human couriers across an entire network in the near future. Weather conditions, local regulations, and customer adoption will continue to dictate where automation is practically viable. Expansion into multiple countries introduces further complexity due to widely varying infrastructure and legal frameworks.

Industry projections indicate rapid growth for last-mile delivery robotics, but these market figures provide limited practical guidance for operators. A more pressing question is whether automation can reduce the cost per delivery without creating new points of failure. The answer depends less on overall market size and more on reliable performance in active, unpredictable settings.

Viewed through an enterprise lens, the Infermove acquisition is not merely a wager on robotics as a product sector. It is a strategic maneuver to strengthen the connection between artificial intelligence, data, and physical operations. For platform companies whose foundations are logistics and mobility, this deeper integration could become a decisive factor in sustaining growth amid persistent cost pressures.

See also: The Law Society: Current laws are fit for the AI era

Explore insights on AI and big data from industry leaders? Visit the AI & Big Data Expo held in Amsterdam, California, and London. This comprehensive event is part of TechEx and co-located with other premier technology gatherings. Click here for further details.

AI News is powered by TechForge Media. Discover additional upcoming enterprise technology events and webinars at this link.