TLDR: NTT and NTT DOCOMO have jointly developed a new generative AI technology called the Large Action Model (LAM). This AI is designed to revolutionize one-to-one marketing by accurately predicting customer intent through the analysis of diverse time-series behavioral data, enabling highly personalized customer experiences.
TOKYO, November 12, 2025 – Telecommunications giants NTT, Inc. and NTT DOCOMO, INC. today announced the establishment of a groundbreaking artificial intelligence technology named the Large Action Model (LAM). This innovative generative AI is poised to significantly accelerate and enhance 1-to-1 marketing strategies by providing unprecedented levels of personalization and customer intent prediction.
LAM operates by analyzing time-series data, organized in the ‘4W1H’ format (Who, When, Where, What, and How), collected from a multitude of customer touchpoints, including both online channels and physical retail stores. Its core capability lies in predicting customer intent with high accuracy, thereby enabling marketing efforts to be precisely tailored to individual customer needs.
The development addresses a long-standing challenge in personalized marketing: the integration and analysis of sequential behavioral data from various customer touchpoints. Historically, the differing frequencies and formats of data across these touchpoints – such as high-frequency app usage logs versus lower-frequency in-store purchase data – have made comprehensive analysis technically complex. LAM overcomes this by specializing in time-series data that encompasses both numerical and categorical information.
Technologically, LAM boasts a structure similar to large language models (LLMs) and utilizes a Transformer-based model for future behavior prediction. Its learning efficiency is further bolstered by innovations like a hierarchical Transformer structure, which intelligently aggregates data of varying frequencies and formats. This allows the model to distinguish contextual differences, leading to a deeper understanding of customer behavior and more accurate predictions.
Also Read:
- Deutsche Telekom Launches AI-Powered RAN Guardian Agent for Autonomous Network Optimization
- TIME Magazine Introduces AI Agent to Revolutionize Reader Engagement
NTT was primarily responsible for the research and development of LAM, while NTT DOCOMO contributed its expertise in customer journey analysis. The integration of these respective technologies is expected to improve the overall efficiency of data utilization across marketing initiatives, ultimately leading to more effective and personalized customer engagement.
