%201.webp)
Flexxon has secured CyberCall funding to develop X-PHY AI Armour, marking the Singapore cybersecurity company's third selection under the programme.
X-PHY AI Armour is being designed as a hardware-rooted secure Model Context Protocol gateway to protect businesses using AI agents. It is intended to sit between an AI agent and the systems it can access, checking requests and applying hardware-based authorisation to high-risk actions.
The project was funded under the "Cybersecurity for AI" challenge in Singapore's Cybersecurity Industry Call for Innovation, known as CyberCall. The programme is run by the CyberSG Talent, Innovation and Growth Collaboration Centre, a joint initiative of the National University of Singapore and the Cyber Security Agency of Singapore.
The technology is aimed at a growing weakness in AI deployments as agents are given more autonomy to handle credentials, query databases, call tools and interact with other software systems. In that model, security controls often rely on software running on the same host environment an attacker may try to compromise.
Under the proposed design, AI Armour would inspect requests at the protocol level for signs of injected instructions or hijacked tools. High-risk operations would then be subject to authorisation anchored in hardware rather than software alone. Audit records would also be written to hardware, so an attacker controlling the host system could not alter the decision or erase the record.
The funding comes as AI agents take on a larger role in enterprise software. Gartner has forecast that 40 per cent of enterprise applications will include task-specific AI agents by the end of 2026, up from less than 5 per cent in 2025.
Rising risk
The wider cybersecurity backdrop in Singapore points to growing pressure on organisations already dealing with frequent attacks. The Cyber Security Agency of Singapore's Singapore Cybersecurity Health Report 2025 surveyed 2,005 businesses and non-profits and found that almost nine in 10 had encountered at least one cybersecurity incident during the year.
Another agency report, Singapore Cyber Landscape 2025/2026, warned that agentic AI systems could automate significant parts of cyber attack chains and shorten the time needed to carry out intrusions. It also cited OpenClaw as an example of open-source agentic AI technology abused to breach development pipelines at scale.
That threat model differs from more familiar software security problems because an AI agent can carry out harmful actions using valid credentials while appearing to operate within normal permissions. A poisoned instruction inserted into a document, webpage or another agent's response can redirect the agent, while the resulting actions still appear authorised to many conventional controls.
Flexxon argues this leaves a gap for businesses in sectors such as financial services, healthcare, government and data centre operations, where agents may be connected to sensitive systems or valuable data. Generic API gateways typically inspect traffic at the network layer, but they can be reconfigured if the host is taken over, according to the company.
"Agents are useful precisely because they act without waiting for a person, and that is exactly what makes them dangerous. An agent following a malicious instruction is indistinguishable from an agent doing its job. You cannot solve that by asking software to watch itself more carefully - the judgement has to sit somewhere an attacker cannot reach. Our thanks to CSA, the TIG Centre and the judging panel for backing us once again and for the collaboration that makes much-needed solutions like ours possible," Camellia Chan, Co-founder and Chief Executive Officer of Flexxon, said.
Third selection
This is Flexxon's third award under the CyberCall initiative. The company was one of eight selected from 72 submissions in the 2019 round for its X-PHY solid-state drive and later received funding again in the 2021 edition for its server security module.
Those earlier projects focused on securing data at rest and the systems around it. AI Armour shifts that hardware security work towards risks linked to autonomous AI systems, where decision-making and system access can happen at machine speed.
Willis Lim, Executive Director of the TIG Centre, said: "Securing AI agents will be a major cybersecurity challenge in the coming years, and it is exactly the kind of problem CyberCall was established to address. We are pleased to support Flexxon as it builds on its expertise to develop X-PHY AI Armour to tackle this emerging challenge."