AI Trace
Data AnalysisVerified

Reviewed and published by trentmaziarz, May 11, 2026. Discovered and drafted by our automated research pipeline.

Zynga, a subsidiary of Take-Two Interactive acquired in 2022, uses a machine learning system called AutoModel that automatically builds hundreds of predictive models daily across all of its mobile games to personalize player experiences and predict player behavior.

Details

AutoModel was developed and publicly described by Zynga's data science team (originally presented at Spark Summit 2019). The system takes player tracking events from Zynga's data lake as input and outputs propensity scores — predictions of which players are most likely to take specific actions, such as making an in-game purchase or churning. These scores are published to a real-time database and used to personalize games and live services across dozens of titles. The system was described as generating 'hundreds of propensity models every day' and scaling to tens of millions of players across Zynga's game portfolio. As Zynga is now a wholly owned subsidiary of Take-Two Interactive, this practice is part of the Take-Two corporate family.

Products affected

Zynga mobile game portfolio (including Words With Friends, Toon Blast, Empires and Puzzles, and others)

Sources & Evidence

Cite this record

Trace Foundation. (2026). Zynga: Zynga, a subsidiary of Take-Two Interactive acquired in 2022, uses a machine learning system called AutoModel that automatically builds hundreds of predictive models daily across all of its mobile games to personalize player experiences and predict player behavior (data as of 2026-05-11) [Data set record]. AI Trace. https://www.aitrace.org/r/practice/c87fd086-d0c0-44ef-84aa-001d3111ccf4. Accessed October 5, 2026.

Stable link
https://www.aitrace.org/r/practice/c87fd086-d0c0-44ef-84aa-001d3111ccf4
Data as of
May 11, 2026
Last verified
Not recorded

How to cite AI Trace

Have evidence about Zynga's AI practices? Submit a report.

Report a Sighting →