Case Study Deskcasestudydesk.com
TelecommunicationsSourced

Case Study: AT&T cuts call-center AI costs with small language models from H2O.ai

AT&T Case StudySourced & dated by Case Study Desk
Key facts · TL;DR
Company
AT&T
Industry
Telecommunications
Challenge
Classifying 15M annual calls with GPT-4-class models was costly to scale
Headline result
An ensemble of small language models hit 91% accuracy at 35% of the prior cost

Key results

91%
Call-classification accuracy
small-language-model ensemble
35%
Of previous solution's cost
cost reduced to
15M
Customer calls handled annually

The challenge

AT&T handles 15 million customer calls annually, generating a vast volume of recorded, transcribed, and summarized interactions. Classifying these calls against an 80-label system using large models such as GPT-4 was computationally expensive and difficult to scale.

The solution

AT&T distilled large language models into three smaller fine-tuned open-source models, including H2O.ai's Danube (a compact 1.8B-parameter model fine-tuned with H2O LLM Studio). The final ensemble combined 10 categories from Llama, 20 from Danube, and 50 from a classifier.

The results, in context

The ensemble of small language models achieved 91% accuracy, closely matching the performance of the previous, more expensive solution. By combining categories across Llama, Danube, and a classifier, AT&T reduced costs to 35% of the previous solution, with Danube accounting for just 10%.

Products used

H2O.ai H2O DanubeH2O.ai H2O LLM Studio