ASEAN leads world in AI observability, New Relic says

New Relic has published its 2026 Observability Forecast for Asia-Pacific, with ASEAN findings showing the region leads the world in AI observability adoption.

Data from Singapore, Indonesia, Malaysia and Thailand also showed that engineering teams in the region spend a median 47% of their capacity dealing with system disruptions, compared with 37% globally. This is despite ASEAN recording the highest use of AI-assisted troubleshooting among the groups surveyed.

The report found that 66% of ASEAN organisations have deployed AI application observability, versus 47% globally. It also found that 59% use AI-assisted troubleshooting, the highest rate recorded in the survey.

Spending plans reflect that focus. In ASEAN, 39% of respondents cited agentic AI adoption as the main driver of observability spending, making it the region's top reason for investment and the highest share among all groups surveyed.

At the same time, the study pointed to rising operational strain as technology environments grow more complex. Nearly half of respondents in ASEAN, or 47%, said tech stack complexity was the main barrier to achieving full-stack observability.

Outage pressure

Service disruptions appear to be becoming more frequent. The proportion of organisations facing high-impact outages multiple times a day rose to 10% from 1% a year earlier, while those experiencing such outages once a day increased to 11% from 4%.

The financial cost is also substantial. The average high-impact outage was estimated at USD $1.62 million, while the mean time to resolve an incident was 52 minutes.

The figures suggest a gap between the region's investment in AI-related monitoring and its ability to resolve incidents quickly. Although ASEAN had the highest uptake of AI-assisted troubleshooting, 23% of technology leaders said they still primarily detect critical outages through manual checks at set intervals, the highest share in Asia-Pacific.

That reliance on manual processes may be linked to fragmented tooling. According to the survey, 53% of ASEAN respondents said they strongly or somewhat prefer using multiple point tools, which can leave telemetry spread across separate systems and make cross-stack correlation harder during an incident.

Standards uptake

The research also found strong OpenTelemetry adoption in ASEAN. A third of organisations, or 33%, said they were fully standardised on the data framework, compared with 22% globally, while a further 27% said they were actively migrating.

Even so, standardised data formats have not removed the operational challenges created by disconnected monitoring tools. Separate systems can keep data in functional silos, leaving engineers to piece together isolated alerts when services fail.

New Relic surveyed 2,575 IT and engineering leaders and practitioners across 24 countries in the Americas, Asia-Pacific and Europe, covering 12 industries. The ASEAN sample included 150 respondents from Indonesia, Malaysia, Singapore and Thailand across executive, management and practitioner roles.

One finding pointed to the growing effect of AI-generated software on monitoring needs. In ASEAN, 86% of leaders said they agree or strongly agree that the rise of AI-generated code makes robust observability more critical, the highest level of agreement recorded globally.

Singapore's policy backdrop was cited as part of the regional context for AI adoption, with the city-state's model governance framework for agentic AI highlighted as a supportive factor for organisations treating AI agents as a strategic priority.

Rob Newell, Senior Vice President and General Manager for APJ at New Relic, said the region had taken an early lead but was now confronting the operational cost of that shift.

"Globally, ASEAN has formed a massive head start in deploying AI system monitoring and application observability," said Newell. "But with engineering teams losing nearly half of their capacity to system firefighting, local innovation is hitting a wall. To fuel the region's multi billion dollar digital economy, leaders must liberate their developers from reactive troubleshooting to focus on building the next wave of disruptive software."