Singapore's military to explore quantum computing for mission planning

The race to be AI-first

The defence deal was announced at IBM Think on Tour Singapore 2026, where Big Blue put out a blueprint for Asia-Pacific organisations to become AI-first enterprises by 2030 – and made the case that quantum readiness belongs on that agenda now.

In her keynote, Ana Paula Assis, IBM’s senior vice-president and chair for Europe, Middle East, Africa and Asia-Pacific, claimed quantum advantage would arrive this year, pointing to the quantum-centric supercomputing reference architecture IBM published in March, the more than 80 quantum systems it has built since 2016, and some 300 organisations already accessing its quantum computers over the cloud.

“Quantum is no longer a scientific problem; it is an engineering problem,” she said. “When something moves from science to engineering, the question is no longer whether it’s going to happen. It’s about when and how fast.”

Assis likened the state of AI adoption to that of electricity: early factories installed light bulbs, but it was the electric motor and assembly line it enabled that transformed production, noting that most organisations are stuck in the "light bulb phase" of drafting documents and automating individual tasks with AI. “The question is no longer ‘can we build with AI?’, but ‘can we operate it at scale, safely, reliably?’.”

IBM used the event to provide its answer – an AI technology stack spanning the next generation of Watsonx Orchestrate for multi-agent orchestration; the Concert platform for intelligent operations; Bob, an agentic AI software development assistant; and IBM Sovereign Core for operational independence.

Hans Dekkers, general manager of IBM Asia-Pacific, weighed in on growing interest in digital sovereignty across the region, noting that sovereignty rests on four connected elements: localised data, open source foundations, local skills, and local governance.

To that end, IBM is working with Visionbay.ai, the supercomputing and cloud AI arm of Hon Hai Technology Group – the Taiwanese contract manufacturer better known as Foxconn – to build a full-stack sovereign AI platform. Targeted at enterprises and regulated industries, the platform will be based on the Red Hat OpenShift container orchestration platform and include development and governance capabilities from IBM's Watsonx portfolio.

Jesse Chao, head of AI and quantum computing for corporate business development in Foxconn's chairman's office, said the need for digital sovereignty will span across the “data supply chain”.

“Whether it’s emergency services or hospitals, you want ownership of your data. You want control, and you want to fit it into your own models,” he said, tipping financial services, defence and telecoms as the first adopters of Foxconn’s sovereign AI platform.

Closing the enterprise gap

Setting the AI policy backdrop in a fireside chat with IBM’s Assis, Josephine Teo, Singapore’s minister for digital development and information, described the city-state’s refreshed national AI strategy as “a double click rather than a system reboot” – one aimed at closing the gap between individual and enterprise adoption.

“Individual improvements in productivity are happening at a faster clip than improvements in productivity at the enterprise level,” she said. To close the gap, Singapore has formed national AI missions to transform the advanced manufacturing, healthcare, finance and connectivity sectors, which together contribute 40% of the country's GDP.

Trust in agentic AI emerged as the event’s most sober thread. Teo warned that existing safety regimes were built for human error, but with AI agents, the errors, risks and safety issues that arise are not completely well understood.

On the customer panel, Guo pointed out that AI governance should scale with the use case – light-touch oversight for low-risk information retrieval, but strict human-in-the-loop controls where AI agents could trigger consequential actions, such as in military operations.

“Regardless of use case, human centricity and human accountability must always be there,” he said. “You cannot hand off an agent just after it has finished development. Responsibility continues. Accountability continues.”

That human-first stance was echoed in the event’s healthcare showcase, which IBM and Changi General Hospital (CGH) described as the first AI-powered training platform for paediatric mass casualty incidents.

Built with Watsonx.ai for real-time scenario intelligence and Watsonx Orchestrate for agentic workflow orchestration, the mixed reality platform features an AI assessor that analyses voice communications, treatment actions and simulation data to give clinicians objective feedback on the human factors – communication, coordination and workflow – that conventional drills struggle to measure.

“The human decision always comes first,” said Jimmy Goh, senior consultant at CGH’s department of emergency medicine, when asked what happens if doctors disagree with the AI. “We will never reach a stage where we allow the decision-making to be taken over by AI, because if that happens, we lose autonomy as human caregivers.”