Philippine Airlines resolves almost half of customer contacts with AI
Philippine Airlines runs its contact centre on Twilio Flex with generative AI chat, and says about 45% of customer concerns are resolved without a live agent. Its customer experience chief also describes predictive models applied to aircraft sensor data and maintenance reports. All figures are the airline's own.
The operating picture
Customer service rebuilt around automation
Aircraft of Philippine Airlines 23. Photograph: Gdfrdymldo1999 / CC BY-SA 4.0 (cropped to fit). Illustrates the airline, not an AI deployment.
PAL carried more than 16 million passengers in 2025, according to Twilio's case study. Mark Anthony Munsayac, vice president for customer experience, has described a deliberate move to digital service, with generative AI chat for online passengers and voicebots being tested on selected call types.
The airline set a target of a "super AI agent" handling about 80% of what a live agent does by April 2026. Munsayac has also described predictive maintenance work on aircraft data; that record rests on his account in a supplier case study and stays unverified in the wider register.
Selected implementations
The work behind the evidence
01 / Deployed
Almost half of concerns resolved without an agent
The operational problem
A network airline takes enquiries across voice, chat and social channels, and long waits on any of them cost goodwill with passengers.
What the evidence reports
Twilio and CMOtech Asia reported in February 2026 that PAL runs its contact centre on Twilio Flex, integrated with airline systems, and has introduced generative AI chat for online passengers. PAL reports average waits under one minute, customer satisfaction around 95%, about 45% of concerns resolved without a live agent and about 30% monthly savings in customer service. At the time AI handled about 10% of customer service tasks. "We're not hiding our live agents," Munsayac said, adding that passengers can always choose to connect with one.
Operational significance
Waiting time and resolution without an agent are the measures passengers feel. PAL publishes both, along with a candid figure for how much of the work AI was doing: about a tenth.
Evidence still needed
All figures come from PAL through its supplier's case study and trade coverage, with no published method or baseline. The sources reviewed do not say whether the April 2026 target of 80% was reached.
A source can verify an announcement without verifying its projected outcome.
The wider register
Other tracked activity
Other records are shown separately, including non-AI technology and programmes at different stages of deployment. A register entry does not establish a measured AI benefit.
AI aircraft predictive maintenance at PAL, vendor-reported · Reported · unverified
Twilio's undated case study quotes PAL as saying it has deployed operational AI, including aircraft predictive maintenance.
This is reported by Twilio and has not been verified.
A source can verify an announcement without verifying its projected outcome.
Editorial assessment
PAL has been open about its customer service numbers, including the modest share of the work AI handled at the start.
The next update should report whether the 80% target was met, and give a fuller account of the predictive maintenance work, which so far rests on one executive's description.
Live data from the Aerostack register. Status reflects published evidence only. See Our approach for how records are graded.
Aerostack register · 2 tracked deployments
Aerostack tracks AI deployments across airports and airlines, with the published evidence behind each record. Photography is real aviation imagery. Record thumbnails may show the named airport or airline; other photographs do not.
