PLDT has deployed three interconnected agentic AI systems across customer service, knowledge management, and risk operations using the UiPath Platform, fundamentally reshaping how the telecommunications operator handles information retrieval and decision-making. KAI compresses knowledge research from five days to one to three seconds, eliminating 25,000–30,000 hours of manual work annually; Ellie reduces enterprise support response times by up to 80% whilst cutting manual effort by 25–35%; ERICA accelerates risk assessments from two to ten days down to five minutes to one day, with 97–99% reduction in manual labour. The scale of these gains—particularly ERICA's near-total automation of risk scoring and treatment recommendations—signals that agentic AI has moved beyond incremental efficiency gains into wholesale process redesign. For CX teams already managing Zendesk or Freshdesk implementations, PLDT's deployment raises a critical question: are your current platforms equipped to integrate with autonomous agents that operate across knowledge bases, workflows, and backend systems simultaneously, or are you still treating AI as a bolt-on layer to existing ticketing infrastructure?
The implications extend beyond labour displacement. PLDT's architecture demonstrates that the competitive advantage now lies in orchestration—connecting disparate data sources (approved knowledge repositories, enterprise workflows, risk registers) into a unified intelligence layer that agents can query and act upon. This mirrors the broader industry shift toward specialised AI agents purpose-built for specific business functions, where generic chatbots give way to role-specific intelligence. For support leaders, this means the next generation of CX tooling will demand tighter integration between your knowledge management systems, CRM, and automation platforms—siloed tools will become liabilities. The 80% reduction in response times and 97–99% automation of routine assessments also reframes what "support" means: agents are increasingly freed from information retrieval and low-complexity decisions, forcing teams to reconsider hiring profiles, training priorities, and what constitutes value-added human work.
PLDT's partnership with UiPath rather than a native CX platform vendor also signals a strategic choice: orchestration platforms may increasingly compete with traditional CX suites for enterprise automation budgets. For mid-market support teams, this creates a vendor landscape question—should you invest in deeper integration with your existing CX platform's AI capabilities, or evaluate whether a horizontal automation platform offers better long-term flexibility? The answer likely depends on whether your organisation's bottlenecks are primarily within customer-facing channels or distributed across enterprise knowledge, workflows, and risk operations. PLDT's three-system approach suggests the latter is increasingly common.
PLDT taps agentic AI to automate customer service, risk management Newsbytes.PH