Grab, best known for ride-hailing, food delivery and digital payments across Southeast Asia, is moving into AI infrastructure. The company has committed $750 million to two AI data centre projects in Malaysia, the country where it was founded.
The investment is part of a larger $1.85 billion joint commitment with SoftBank, one of Grab's major shareholders, and Malaysian state-owned oil and gas company PETROS. According to a report by Asia Tech Review, the backers say the projects are intended to become one of Asia's largest AI infrastructure developments once operational.
How the $1.85 Billion Is Divided
The money is split across two sites. The first, in Peninsular Malaysia, will receive $650 million from Grab and $200 million from SoftBank.
The second project, in the state of Sarawak, is led financially by SoftBank, which is contributing $800 million. Grab is adding $100 million, and PETROS is putting in another $100 million.
Taken together, Grab accounts for $750 million of the total, SoftBank for $1 billion and PETROS for the remaining $100 million. The partners are aiming for at least 5GW of capacity across the initiative.
Seed Money With Room for More Investors
The initial capital is being described as seed funding rather than the full cost of the build-out. The founding backers appear to expect additional investment from outside parties as the projects develop.
Grab CEO Anthony Tan will oversee both projects, while a separately appointed management team will handle day-to-day operations. That structure suggests the data centres are intended to function as distinct businesses rather than internal Grab facilities.
Ties to Malaysia's National AI Strategy
The participation of PETROS signals that the effort is closely linked to the Malaysian government and fits within the country's national AI plan. Malaysia has already emerged as a leading data centre hub in Southeast Asia, much of it built to serve large US cloud providers.
These projects go a step further by focusing specifically on AI computing with government involvement. Separately, Malaysia is planning a dedicated wholesale electricity market for large power users such as data centres by 2029 or 2030, a sign of how central energy supply has become to the sector.
Tan framed the investment as a way to build AI tailored to the region, saying the infrastructure will help Grab "accelerate the development of localized AI solutions for our communities." He also pointed to high-value job creation and AI skills training as benefits for the Malaysian economy.
Why Grab Can Afford the Bet
Grab has substantial resources to deploy. It currently holds $2.85 billion in cash and cash equivalents, even after an active year of dealmaking.
This year the company acquired fintech startup Stash and agreed to buy lending business Atome for an initial $1.5 billion. A planned $600 million purchase of Foodpanda's Taiwan operations recently fell through. The AI infrastructure commitment marks a new direction for that spending.
Open Questions and Investor Reaction
It remains unclear how much of the new capacity Grab will use for its own services. If the projects attract further funding, Grab could eventually earn returns that offset its outlay while gaining the strategic advantages of owning computing infrastructure.
The move stands out for a company that has often been seen primarily as a transportation business, and one that has faced questions about its ambitions in technology and fintech. Regional rival Sea, by contrast, has been expanding into cloud computing services.
Investors gave a cautious first response. Grab's share price fell 2.5% on the announcement, though the stock had risen 6% over the previous month following strong quarterly results that showed both profitability and growth.
What the Move Signals for Southeast Asian Tech
Grab's commitment shows that the region's largest consumer tech platforms increasingly view AI infrastructure as strategic territory, not just something to rent from global cloud providers. Returns on projects of this scale are likely to take years to materialise.
For founders and investors in Southeast Asia, the partnership highlights the growing role of governments, state-linked companies and deep-pocketed backers like SoftBank in shaping where AI capacity is built. Whether outside capital follows will be an early test of how far these ambitions can go.
Source: Asia Tech Review
