Key facts
- xAI has filed its first lawsuit against a user, Terry Wayne Harwood, accused of generating child sexual abuse material (CSAM) with its Grok chatbot.
- The lawsuit alleges Harwood used multiple xAI accounts to create non-consensual sexualized images of victims, including a young girl.
- xAI's legal action aims to hold users solely liable for illegal content generated by the AI, citing breach of contract and terms of service violations.
- This move comes amid pressure on xAI to address concerns about Grok's potential for generating harmful and illegal content.
- The lawsuit could impact a proposed class-action lawsuit where xAI is accused of not adequately assisting law enforcement in identifying users of its platform.
Elon Musk's xAI has filed its first lawsuit against a user, Terry Wayne Harwood, accused of generating child sexual abuse material (CSAM) through its Grok chatbot. The lawsuit, filed in a U.S. district court, seeks to establish that users, not the AI company, are liable for illegal content produced via the platform.
Harwood was previously arrested for possession and distribution of CSAM. xAI alleges that he used at least two accounts between December 8 and February 18 to create non-consensual sexualized images of multiple victims, including a young girl. The company claims Harwood violated xAI's terms of service, which prohibit generating intimate or sexual content involving real people's likenesses or depicting children in a pornographic manner.
The lawsuit comes as xAI faces mounting pressure regarding Grok's capabilities to produce harmful content. A separate proposed class-action lawsuit alleges that xAI has not adequately assisted law enforcement in identifying users who upload illegal material. In that case, a victim claims her stepfather created thousands of sexualized images of her using Grok and distributed them online.
xAI argues in its complaint that Grok should be considered a neutral tool, with all outputs being the result of user prompts and directions. By suing Harwood for breach of contract, xAI aims to avoid substantial legal fees and potential liability, asserting that users are responsible for all content, including inputs and outputs. The company stated that any CSAM uncovered is reported to NCMEC, though it did not provide examples of Harwood's successful prompts or methods used to bypass safeguards, citing a desire to avoid aiding other bad actors.