customer_support · travel · workflow
How AI text generation models are reshaping customer support at Airbnb
Scaling AI for Airbnb's customer support was difficult due to long-tail corner cases, the high cost of human data labeling, and the limitations of traditional classifiers that could not scale to exhaustive intent taxonomy design.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · User issue submitted to ranker
The content recommendation ranker receives a user's current issue description alongside candidate support documents.
Tools used
MT5XLMRoBERTaDeepSpeedt5-baseNarrativaT5BARTPEGASUSGPT2Sentence-Transformers
Outcome
Airbnb's generative AI models significantly improved content recommendation ranking performance, drove large engagement rate improvements for CS ambassadors using the agent assistant, and significantly improved chatbot engagement rates through paraphrasing.
What failed first
The content recommendation ranker previously used XLMRoBERTa, which was replaced by a generative model. The initial paraphrase model generated bland, generic replies regardless of the specific user input.
Results
Volumesignificant improvements in the key performance metric for support document ranking
Grounding & classification
Source type: technical build writeup
38 fields verified against source quotes.
content generationconversational aiknowledge searchsummarizationsupport agentchat transcriptknowledge basefailure mode describedhuman review describedmetric backednamed customerproduction runtime claimedsource backedtools describedworkflow describedhospitalitytravelaccuracy improvementcustomer satisfactionemployee productivitytechnical build writeupcall center aicustomer supportai draft human approvalautonomous resolutionrag answering