started · updated
Sony files patent for AI decoy accounts to detect bad actors
Sony Interactive Entertainment has filed a patent application (no. 18/314775) for an AI-powered social integrity system designed to proactively detect bad actors on the PlayStation Network.
The proposed system utilizes large language models (LLMs) to operate “pseudo accounts” that act as decoys. These accounts are designed to mimic the messaging patterns of specific demographics, such as minors, to attract “anti-social” users including scammers, spammers, cheaters, and potential predators.
Rather than relying solely on reactive moderation, the system would flag suspicious accounts based on their interactions with these AI-controlled decoys. Once flagged, the system can monitor the user and issue warnings to legitimate players before any harmful communication occurs.
Entities
Claims
What the coverage asserts, and how many sources carry each claim.
- [● 2 SOURCES] The system provides warnings to real players who interact with flagged accounts. gamingbolt.com · mp1st.com
- [● 3 SOURCES] Pseudo accounts can mimic the messaging behavior of specific demographics, such as minors. www.blogspan.net · gamingbolt.com · mp1st.com
- [● 3 SOURCES] Sony Interactive Entertainment filed patent application 18/314775. www.blogspan.net · gamingbolt.com · mp1st.com
- [● 2 SOURCES] Flagged users can be monitored until they attempt to contact a legitimate player. gamingbolt.com · mp1st.com
- [● 3 SOURCES] The patent aims to identify scammers, spammers, cheaters, and potential predators. www.blogspan.net · gamingbolt.com · mp1st.com
- [● 3 SOURCES] The system uses a large language model (LLM) to control pseudo accounts. www.blogspan.net · gamingbolt.com · mp1st.com