Universitas Gadjah Mada, Indosat Ooredoo Hutchison and Nvidia opened Indonesia's first university-based AI technology center in Yogyakarta this week, giving local researchers access to sovereign GPU compute for the first time. The UGM Indosat NVIDIA AI Technology Center, or NVAITC, was launched on August 14, 2026 under Indonesia's AI Center of Excellence initiative, a joint effort with the Ministry of Communication and Digital Affairs. The center's first three projects target tuberculosis screening, precision agriculture and natural disaster response — problems that hit Indonesia harder than almost anywhere else.
The scale is the story. Indonesia is the world's fourth-most populous country and records over 1 million new tuberculosis cases every year. Nearly 30% of its workforce is employed in agriculture, much of it smallholder farming across archipelago terrain. And the country sits on the Pacific Ring of Fire, facing near-constant seismic and volcanic risk. Until now, researchers working on those problems have lacked domestic access to enterprise-grade accelerated computing.
The center is powered by Nvidia's full-stack AI platform running on GPU Merdeka, Indosat's sovereign GPU-as-a-service offering. That gives UGM researchers and students access to Nvidia Nemotron open models, development frameworks, pretrained checkpoints and technical mentorship without routing workloads through overseas clouds. Sovereignty is the operative word — the compute, the models and the data stay in-country.
“The Indonesia AI Center of Excellence reflects our long-term vision to position Indonesia as a nation that not only adopts AI but also develops and contributes AI innovations to the world.”— Meutya Hafid, Indonesia's Minister of Communication and Digital Affairs
Key facts
- 01UGM Indosat NVAITC opened in Yogyakarta on August 14, 2026 as Indonesia's first university-based AI technology center.
- 02The center runs on Nvidia's full-stack AI platform and Indosat's GPU Merdeka sovereign GPU-as-a-service, with access to Nvidia Nemotron open models.
- 03Indonesia records more than 1 million new tuberculosis cases each year, one of three focus areas for the center's initial projects.
- 04Nearly 30% of Indonesia's workforce is employed in agriculture, targeted by the SmartAgri multimodal AI project.
- 05The launch is part of Indonesia's AI Center of Excellence initiative announced by the Ministry of Communication and Digital Affairs (Komdigi).
Meutya Hafid, Indonesia's Minister of Communication and Digital Affairs, framed the launch as foundational to what she called Indonesia's AI sovereignty, positioning the country as a contributor of AI innovation rather than a consumer of it. The government is treating AI infrastructure the way it treats other strategic assets: something that has to sit on Indonesian soil, run on Indonesian terms and serve Indonesian priorities first.
Indosat CEO Vikram Sinha said the operator's role is to widen access across the research, student, startup and developer ecosystem, not just fund a single lab. GPU Merdeka is Indosat's bet that a domestic telecom can serve as the country's AI compute backbone, competing with hyperscaler regions that don't yet have a footprint dense enough to serve Indonesian researchers at low latency.
Marc Hamilton, Nvidia's vice president of solutions architecture and engineering, cast the partnership as talent enablement — pairing Nemotron open models with local expertise to build products that work for Indonesian conditions. Nvidia has been steadily seeding AI technology centers in emerging markets, and the Yogyakarta location gives it a foothold in Southeast Asia's largest economy at the university level, where the next generation of researchers is trained.
“At Indosat, we believe no Indonesian should be left behind in the AI era.”— Vikram Sinha, President director and CEO of Indosat Ooredoo Hutchison
The tuberculosis project is the most concrete. Led by dr. Dian Kesumapramudya Nurputra of UGM's Faculty of Medicine Public Health and Nursing, the eNose-TB system uses AI to screen for tuberculosis from breath samples. TB is curable when caught, but detection in rural clinics has required lab equipment and specialists that most Indonesian villages don't have. An accurate breath-based screener that runs on portable hardware would collapse that access gap — the difference between a treatable case and a fatal one is often just whether someone got tested.
The two other flagship projects follow the same pattern of applying AI to specifically Indonesian constraints. SmartAgri combines satellite imagery, sensor data and local agricultural knowledge with edge computing to deliver irrigation and planting recommendations tuned to Indonesian crops and terrain — a market where nearly 30% of workers depend on the outcome. Tech4Disaster processes satellite and sensor data on Nvidia accelerated compute to give emergency responders faster warning and situational awareness when earthquakes, tsunamis or eruptions hit.
“As researchers, we have always believed that technology developed in Indonesia can solve Indonesian challenges.”— dr. Dian Kesumapramudya Nurputra, UGM Faculty of Medicine Public Health and Nursing
The open question is throughput. A single university-based center — even one with sovereign GPU access — will not by itself produce the volume of trained AI engineers Indonesia needs to match its ambitions, and the announcement did not disclose GPU counts, seat numbers or a target for how many researchers the facility can support at once. Whether Indosat's GPU Merdeka platform can scale to serve the broader ecosystem Sinha described, or whether NVAITC becomes a bottleneck itself, will depend on capacity expansions the partners haven't yet detailed.
For Nvidia, the strategic logic is straightforward: locking in Nemotron and CUDA as the default stack at the university that will train Indonesia's next decade of AI researchers is worth more than any single enterprise deal. For Indosat, it turns a telecom into an AI infrastructure vendor at the exact moment governments across Southeast Asia are writing sovereignty requirements into procurement. And for Indonesia, it's the first serious attempt to move from AI adoption to AI production — with 280 million people and problems worth solving, that shift is where the real economic upside sits.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
