flydubai buys AI for turnarounds, crew immigration and customer data
Between July 2025 and January 2026 flydubai announced four technology agreements described as AI-driven. Only one, biometric immigration gates for its crews, is confirmed as installed. The others are supplier selections and a framework agreement, none with a published start date.
The operating picture
Four agreements in seven months
flydubai Boeing 737. Photograph: N82khan / CC BY-SA 4.0 (cropped to fit). Illustrates the airline, not an AI deployment.
flydubai runs a Boeing 737 fleet from Dubai International (DXB), where it operates about 370 flights a day by its own count. Its recent purchases target that hub operation directly: getting aircraft turned round on time, clearing crews through immigration before duty, and recognising passengers across its systems.
The airline's own releases are the primary sources, and each quotes a named executive. They are clear about what has been selected and far less clear about what is running. A December 2025 framework with Amazon Web Services covering generative AI names no use case, and a Panasonic inflight entertainment order makes no AI claim; both stay in the wider register.
Selected implementations
The work behind the evidence
01 / Contracted
Watching every turnaround at DXB
The operational problem
With about 370 flights a day at its hub, a delayed task on one aircraft can hold a departure and push delay into the next wave of flights. Ground teams need to see the problem forming while there is still time to recover.
What the evidence reports
flydubai said on 9 January 2026 that it had selected ZestIoT to deploy a turnaround management platform at DXB and across its network. The release describes real-time visibility of turnaround activity and "AI-driven and prediction-based collaboration". Mohamed Hassan, senior vice president of airport services and cargo, referred to an existing "successful partnership" with the supplier.
Operational significance
Turnaround performance is measured in minutes against records the airline already keeps, such as on-time departures and actual turnaround times. That makes this the most testable of flydubai's agreements.
Evidence still needed
The release does not say what the system predicts, how it captures turnaround events, which stations follow DXB or when deployment will be complete. The earlier partnership it mentions is not described.
Pilots and cabin crew clear immigration before every duty. flydubai has more than 1,300 pilots and 2,500 cabin crew reporting through its Airport Operations Centre.
What the evidence reports
flydubai announced in July 2025 that it had installed six emaratech smart gates at the Airport Operations Centre, linked to the airline's systems and UAE immigration platforms. emaratech's chief executive described the technology as AI-powered facial recognition, and NEC said in December 2025 that its NeoFace facial recognition units power the gates.
Operational significance
This is the only flydubai record with equipment confirmed in place. Facial matching is an established use of machine learning. The operational change is paperless crew clearance linked directly to immigration.
Evidence still needed
Neither the airline nor its suppliers have published clearance times, throughput or error rates, and the AI description comes from the suppliers.
Booking, check-in, inflight and customer service systems each hold their own record of the same traveller, which limits recognition and relevant offers.
What the evidence reports
flydubai announced on 12 January 2026 that it had adopted Amperity's customer data platform, which links passenger records using what Amperity calls AI-driven identity resolution. The airline said it would use the platform across communications, check-in, inflight recognition, customer service and marketing. Mohammed Hareb AlMheiri, chief procurement and technology officer, said passenger data is handled under the Dubai Electronic Security Centre's information security framework.
Operational significance
Identity resolution is plumbing. It decides whether later personalisation works, because an offer is only relevant if the system knows that two bookings belong to the same person.
Evidence still needed
The release is written in the future tense and gives no start date or measure of how accurately records are matched.
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.
flydubai and Amazon Web Services collaboration on generative AI, data and machine learning · Contracted
flydubai and Amazon Web Services announced a collaboration on 15 December 2025, signed in the presence of Mohammed Hareb AlMheiri, flydubai's Chief Procurement and Technology Officer, and Werner Vogels, Amazon's Chief Technology Officer. The airline says the agreement sets up a framework of structured innovation engagements in generative AI for airline operations, data analytics, machine learning and customer experience.
The stated aim is to develop and scale digital and AI use cases across operations and the customer journey and to speed the delivery of production-ready systems. No specific use case, term or value was named, and no production system from the collaboration had been announced as of 16 October 2026.
Panasonic Astrova inflight entertainment and Converix for flydubai's 30 Boeing 787-9s · Contracted
flydubai has selected Panasonic Avionics' Astrova inflight entertainment system for its 30 Boeing 787-9s on order, the airline announced on 19 November 2025. The deal also covers Converix, which Panasonic says will help flydubai increase digital engagement and loyalty and improve operational efficiency.
flydubai says Astrova will let it offer richer content, smarter personalisation and a more intuitive interface, and Panasonic highlights 4K screens. The system will enter service when the 787 joins the fleet. No first-delivery date for the 787-9s or details of how personalisation will work were given as of 16 October 2026.
flydubai has moved quickly from one agreement to the next, and its own releases put a named executive behind each. Delivery is the open question: one installation is confirmed, and the turnaround platform, the most operationally significant of the four, has no published completion date.
The next update should establish whether ZestIoT is running beyond DXB, and whether flydubai reports any change in turnaround or on-time performance that it attributes to the system.
Live data from the Aerostack register. Status reflects published evidence only. See Our approach for how records are graded.
Aerostack register · 5 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.
