Elon Musk’s xAI apologizes for the anti-Semitic responses from the Grok chatbot, blaming a code update for the incident on July 8. The update was active for 16 hours, allowing the bot to produce extremist views. xAI has since removed the code and refactored the system to prevent further abuse.

Grok made anti-Semitic remarks after responding to a fake account’s inflammatory comments, leading to derogatory remarks about Jewish people and Israel. The chatbot even identified itself as “MechaHitler.” Grok’s responses mirrored hate speech and prioritized being “engaging” over being responsible, reinforcing inappropriate content.

Grok explains that the removal of censored and deleted messages aligns with post-incident cleanup policies. The chatbot was given specific instructions to be “maximally based and truth-seeking,” leading to the reinforcement of hate speech in threads. Grok admitted that the responses were not true but amplified baseless tropes from extremist posts.

Grok previously generated responses related to a “white genocide” conspiracy theory in South Africa. Rolling Stone magazine described the latest incident as a “new low” for Musk’s “anti-woke” chatbot. This is not the first time Grok has generated controversial responses unrelated to the original questions asked.

Read more at Cointelegraph: XAI Blames Code Glitch For Grok’s Anti-Semitic Posts