Cisco warns Asia Pacific firms lack confidence on AI agents

Cisco has released research showing that many organisations in Asia Pacific, Japan and China lack confidence in dealing with compromised AI agents. In Singapore, confidence is even lower than the regional average.

Its Relentless Defence Report found that only 52% of organisations across Asia Pacific, Japan and China were confident in detecting and responding to a compromised AI agent. In Singapore, the figure was 38%.

The findings come as cyber incidents remain widespread. Across Asia Pacific, Japan and China, 91% of organisations suffered a material, business-disrupting cyber incident in the past 12 months, compared with 89% in Singapore. More than one in three of those incidents in the wider region involved an AI-enhanced attack, while in Singapore the share was more than two in five.

Operational gaps

The report argues that the main constraint on cyber defence is organisational rather than technical. Around six in 10 practitioners globally who said they were struggling pointed to coordination, approvals and access to data as the changes that would most improve their defences.

Across both the regional and Singapore datasets, only 39% of organisations said security, networking and IT teams operate as one team with shared visibility and tooling. That means most still work in separate structures, even as AI systems increase the pace and scale of security threats.

In Asia Pacific, Japan and China, 43% said the metrics they track are too technical to persuade executives, 38% said they struggle to translate business metrics into information that can influence decision-making, and 30% said they have no consistent way to measure security performance. In Singapore, the comparable figures were 42%, 34% and 30%.

Data access also emerged as a recurring obstacle. In the wider region, 42% of organisations said they spend more time collecting and correlating data across systems than addressing the threat itself, while 41% said critical insights are missed because data is out of reach. In Singapore, 37% said they spend more time gathering and correlating data than tackling threats, and 41% reported missed insights for the same reason.

AI adoption

The research suggests AI is improving parts of cyber defence even as concerns about its use remain unresolved. More than eight in 10 practitioners across Asia Pacific, Japan and China reported detection and response gains of at least 25%, and more than one in four said AI was significantly reducing operational burden.

In Singapore, nearly eight in 10 practitioners reported gains of at least 25%, while one in five said AI was significantly reducing operational burden.

Yet deployment remains limited. Fewer than half of organisations in both markets have extensively deployed AI-powered security tools. In the wider regional survey, adoption was highest in data security and privacy and network security, and lowest in security operations response and investigation. In Singapore, adoption was highest in security operations detection and network security, and lowest in AI workloads, models and systems.

Use of AI agents in security operations was particularly limited. Just one-third of organisations in Asia Pacific, Japan and China said they had extensively deployed them, while in Singapore the figure was just over a quarter.

Security concerns were a central reason for caution. Organisations in the wider region cited worries about data exposure, compromised systems and third-party handling, while Singapore respondents pointed to data exposure as well as visibility into and access to AI systems.

Top performers

The report identified a group it calls "Relentless Defenders", defined as the top 8% globally. These organisations accounted for 7% of respondents in Asia Pacific, Japan and China and 6% in Singapore. The average score for both the regional and Singapore samples was 63 out of 100.

Among organisations that increased security spending, this top-performing group was nearly twice as likely to report a fall in incidents. The figure was 71% for the group, compared with 38% in both Asia Pacific, Japan and China and Singapore.

The strongest performers were more likely to organise security, network and infrastructure teams as a single coordinated function with shared visibility and tools. They were also more likely to apply consistent controls to AI agents and non-human identities, and to report faster improvements in threat detection.

Ben Dawson, President, Asia Pacific, Japan and Greater China, Cisco, said many organisations now face a broader question about whether their systems and processes are ready for AI deployment. "Boards have moved past asking whether AI will deliver value. The urgent question leaders are now facing is whether their digital foundation is secure and resilient enough to deploy it. Organisations that treat security merely as a compliance exercise will struggle, while those embedding defense directly into how they operate are the ones setting the pace for innovation in the AI era," Dawson said.

For Singapore, Cisco linked the findings to a need for faster decision-making across internal teams. "AI has fundamentally raised the stakes for cyber defense. In a post-Mythos era, the speed of defense is driven by operational agility, not technology alone. The organizations that stay ahead will be those that embed security into how they operate, treat it as more than a compliance exercise, and break down operational silos so the right people can make the right decisions at the right time," Tay said.