Delivering faster airline accountability with AI

Challenge: The number of air travel complaints increased quickly during the pandemic, growing from about 15,000 in 2019 to more than 100,000 in 2020. Complaints have since leveled off, but remain much higher than before the pandemic, averaging around 75,000 a year. Early reporting from 2026 shows that pace continuing. A prior $8 million TMF investment funded OACP’s Aviation Complaint, Enforcement, and Reporting System (ACERS), replacing a 30-year-old database with a platform that makes it easier for consumers to file complaints, facilitates monthly reporting on complaint trends to the public, and enables airlines to access and respond to complaints in real time. While ACERS improved public intake and real-time carrier access, staff still spend critical hours on manual administrative tasks such as manually searching records, verifying complaint categories, and checking for complaint duplicates, leaving little time for anything beyond what the law requires.

Approach: With TMF support, DOT is integrating secure AI features, including large language models grounded in OACP’s own verified rules and records through retrieval-augmented generation, to:

• Automate complaint categorization and duplicate detection
• Identify trends across complaint categories beyond the statutory minimum
• Pull records faster for congressional, FOIA, and public inquiries

A person will review every automated decision and minimum risk management practices will be followed, as federal rules on oversight require.

DOT expects this investment to:

• Identify complaint trends up to 80% faster, freeing staff to focus on holding airlines accountable
• Achieve 80 to 85% accuracy in automated complaint classification and duplicate detection
• Decrease records request response time by up to 75%, speeding up responses to FOIA and public information requests