NVIDIA Eyes $4.5 Billion Batam Data Centre for National AI Compute
Indonesia’s coordinating minister for economic affairs, Airlangga Hartarto, said on September 26, 2026 that NVIDIA is eyeing a data-centre and national-computing investment of about four and a half billion US dollars in Batam, Riau Islands, as Jakarta pushes special economic zones for artificial-intelligence infrastructure. Local coverage put the figure near eighty trillion rupiah and cast Batam’s land, power, green energy and cooling water as the draw for GPU-heavy halls. Airlangga said Indonesia’s data-centre capacity is roughly eight hundred megawatts today with a pipeline that could approach one point three gigawatts, and he paired the NVIDIA talk with open-source AI sovereignty goals and Indonesia’s founding-member role in China’s WAICO forum awaiting DPR ratification.
Filed under Infrastructure and dated September 26, 2026, this AI4Indonesia briefing treats the Batam pitch as Indonesian hyperscale-AI news distinct from yesterday’s Microsoft 1.7-billion-dollar cloud commitment. Ministers want foreign campuses to double as national computing capacity rather than pure export tenancy, while skills programmes such as Microsoft Elevate’s claim of more than 2.4 million Indonesians trained underscore that racks without talent still stall.
Why it matters: Indonesian startups and agencies still import scarce GPUs. A Batam-scale campus can cut latency—but only if power, residency options and knowledge-transfer clauses are real.
What it means in practice
Indonesian enterprise and investment leads should demand published megawatt and interconnection calendars; map which regulated corpora may enter Batam tenancy; assign an owner for SEZ incentive diligence; run time-boxed migrations of non-critical training jobs; and prefer designs that keep humans on high-impact deployments. Anchor the plan to Microsoft’s Indonesia cloud pledge and Zankore’s AI-factory financing.
Caveats come first. Ministerial “looking at” language is not a signed campus; power and water can slip; and WAICO politics may complicate US vendor optics. AI4Indonesia therefore presents the 4.5-billion-dollar Batam talk as directional infrastructure context until site-level filings appear.
What to watch next: named JV partners and permit timelines; how open-source sovereignty rules interact with NVIDIA stacks; and whether MSME developers get reserved capacity. Readers can continue on the AI4Indonesia homepage, or browse the Newsroom for additional briefings.
Bottom line: treat this update as orientation, not instruction. Indonesian AI infrastructure is courting another hyperscale-scale cheque and remains contingent. Organizations that benefit most will verify megawatts, keep humans on deployment gates, and refuse to confuse a ministerial quote with finished racks.