US 12,138,552 B2
Game analysis platform with AI-based detection of game bots and cheating software
Sebastian Risi, Copenhagen S (DK); and Christoffer Holmgård Pedersen, Copenhagen K (DK)
Assigned to modl.ai ApS, Copenhagen N (DK)
Appl. No. 17/756,206
Filed by modl.ai ApS, Copenhagen N (DK)
PCT Filed Nov. 16, 2020, PCT No. PCT/US2020/060669
§ 371(c)(1), (2) Date May 19, 2022,
PCT Pub. No. WO2021/126436, PCT Pub. Date Jun. 24, 2021.
Claims priority of provisional application 62/949,079, filed on Dec. 17, 2019.
Prior Publication US 2022/0410015 A1, Dec. 29, 2022
Int. Cl. A63F 13/00 (2014.01); A63F 13/533 (2014.01); A63F 13/75 (2014.01); A63F 13/79 (2014.01); G06N 20/00 (2019.01)
CPC A63F 13/75 (2014.09) [A63F 13/533 (2014.09); A63F 13/79 (2014.09); G06N 20/00 (2019.01)] 18 Claims
OG exemplary drawing
 
1. A method comprising:
generating, via a game analysis platform that includes at least one processor and at least one memory, a training data set based on game data collected from actual players;
training, via the game analysis platform, a primary artificial intelligence (AI) model using machine learning based on the training data set to detect player bots;
gathering, via the game analysis platform, actual game data from game play;
processing the actual game data via the primary AI model to generate primary detection results;
detecting, via the primary AI model, a potential use of a player bot when the primary detection results exceed a detection threshold;
training a secondary AI model to recognize actual players; and
in response to detecting the potential use of the player bot via the primary AI model:
processing the actual game data via the secondary AI model to generate secondary detection results; and
confirming use of the player bot when the secondary detection results indicate the use of the player bot.