Airport intelligence / Gimhae International Airport
Gimhae: a customer-service AI pilot inside a much larger airport plan
Korea Airports Corporation tested a generative-AI passenger assistant across its airport network, including Gimhae. A later master-plan contract imagines more ambitious AI uses but does not establish that they are operating.
Gimhae International Airport, Busan. Photograph: 螺钉 / CC BY-SA 3.0 (cropped to fit). Shows the airport, not an AI implementation.
The picture at Gimhae International Airport
What is live, and what is only on the drawing board?
An airport operator managing 14 airports can trial one service across its network without changing every local operation. Korea Airports Corporation's passenger chatbot is an example: an initial Gimpo test was expanded to other airports, including Gimhae, with handoff to a call centre for questions it could not resolve.
In August 2026, LG CNS announced an AI master-plan assignment covering that same network. Its proposed applications included voice-based incident reports and agent assistance for operational safety, but the work was at planning and proof-of-concept stage.
Gimhae also uses palm-vein biometric identity gates. Those are documented airport technology, not evidence that the proposed agentic safety functions are in use.
Selected implementations
An actual service beside a proposed programme
01 / Pilot
A passenger question that can still reach a human
The operational problem
Network airport operators receive routine enquiries about parking, transport and airport services that do not always need an agent.
What the sources report
Korean technology reporting in 2024 described KAC's generative-AI chatbot pilot at 14 airports, with an escalation path to the contact centre. Gimhae was included in the programme.
Why it matters
The handoff to a human matters when the assistant cannot resolve an enquiry.
What remains unproven
Current airport-specific use, accuracy, escalation rate and measured customer benefit have not been published in the attached evidence.
Evidence supports a reported claim; it does not automatically establish a measured operating benefit.
02 / Contracted
An AI safety plan is not yet a safety system
The operational problem
Field staff documenting an incident need fast, accurate reports and a clear operating procedure.
What the sources report
LG CNS announced in August 2026 that it had won KAC's AI master-plan project. The work includes data architecture and proofs of concept; an example agent would draft incident reports from voice descriptions.
Why it matters
The concept could address administrative workload, but safety-critical use would require tight validation and accountability.
What remains unproven
No agentic safety service is confirmed live at Gimhae. The record is contracted, not deployed.
Evidence supports a reported claim; it does not automatically establish a measured operating benefit.
The wider register
What else is happening at Gimhae International Airport?
Gimhae's palm-vein gates are an established biometric service with their own identity and eligibility rules. They belong in the wider register rather than being used to bolster an AI master-plan narrative.
Palm-vein biometric ID gates for international departures at Gimhae · Deployed · limited scope
At Gimhae International Airport, departing passengers can confirm their identity at biometric gates using palm-vein data, Metro Seoul reported on 17 January 2023. The gates have served domestic flights since 2018 and were opened to international departures in December 2022.
Passengers register in advance, and eight unmanned registration kiosks sit in the international terminal. One registration works at all 14 airports run by Korea Airports Corporation, the paper said, and NBN also reported on the Gimhae gates.
Evidence supports a reported claim; it does not automatically establish a measured operating benefit.
The operator-wide plan needs local proof
A common technology programme can make sense across KAC's network, but it still needs airport-level adoption evidence.
For Gimhae, the next meaningful update is whether the passenger chatbot moved beyond pilot status and whether any planned safety application entered controlled use.
Live data from the Aerostack register. Status reflects published evidence only. See Our approach for how records are graded.
Aerostack register · 3 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.