U Mobile deployed the agentic helpdesk AiDaH on the Amazon Bedrock AgentCore platform
Malaysian operator U Mobile, together with AWS and partner Axrail, has launched the agentic helpdesk AiDaH on Amazon Bedrock AgentCore. According to the company, it covers over 150 operational scenarios and reduced the volume of trouble tickets in key categories by up to 84.5%.
Malaysian mobile operator U Mobile, in collaboration with AWS and partner Axrail, has created an agentic helpdesk called AiDaH (AI-Driven Agentic Helpdesk), built on the Amazon Bedrock AgentCore platform. Development began in late 2025 with two pilot agents and gradually expanded: in the first quarter of 2026 it covered the internal IT helpdesk with 4 agents and 20 scenarios, by mid-year it reached 6 agents with three-tier routing, and in the third quarter of 2026 it was deployed for 50 call center and branch users with 8 agents. According to the company, it now covers over 150 documented operational scenarios across six business divisions.
Unlike typical AI assistants, which only help formulate a response, AiDaH, as described in the source, independently goes through multiple U Mobile backend systems (error logs, APIs for verifying customer entitlement, internal rule knowledge bases), analyzes the results, and proposes a specific remediation procedure – for example, which services to reset or whether to log a network ticket. A human employee only reviews and confirms the result. The system runs on a six-layer architecture in the AWS Asia Pacific (Singapore) region, with users accessing it via Microsoft Teams and a web portal; it uses a managed agent runtime without the need to operate its own agent infrastructure, and built-in memory for conversational context without an external state store.
According to U Mobile's internal measurements, the system reduced the volume of trouble tickets in key categories by up to 84.5%. This is a claim made by the company and has not been independently verified. The source does not elaborate further on specific financial benefits or details of the security layer – for details, see the source article.
Why it matters
This case shows a concrete deployment of agentic AI in the operations of a telecommunications company: the system itself goes through multiple internal systems and proposes solutions, rather than merely helping draft the text of a response, while human oversight remains at the point of final action approval. For companies with a fragmented IT landscape (various systems, portals, knowledge bases), it's an example of how a managed agent platform can be used to reduce helpdesk burden without building one's own agent infrastructure – although the reported figure on ticket reduction comes from the company's internal measurements, not an independent audit.
Two audiences, two different impacts
What this means
For individuals
For people in support or IT operations roles, this case shows how the work shifts from manually tracking down the causes of problems to reviewing and approving actions proposed by the agent.
For a business
The company U Mobile replaced manual searching across multiple systems (logs, BSS APIs, knowledge bases) with an agentic system that, according to its internal measurements, reduced the volume of trouble tickets in key categories by up to 84.5% and covers over 150 operational scenarios without the need to build its own agent infrastructure.
ProcessesCheck the original
Event sources
only one source so far · 1 publisher, 0 independent. We count feeds from the same owner only once.