xAI's Grok Ranks Last in ADL Study on Antisemitic Content Moderation

• xAI's Grok performed the worst among six leading large language models (LLMs) in an Anti-Defamation League (ADL) study on countering antisemitic content.
• Anthropic's Claude ranked best, while Grok scored lowest with an overall score of 21, demonstrating consistently weak performance across all categories.
• The ADL tested Grok, ChatGPT, Meta's Llama, Claude, Google's Gemini, and DeepSeek using prompts categorized as "anti-Jewish," "anti-Zionist," and "extremist."
• The study highlighted that all models had gaps and required improvement, with a 59-point spread between Claude's (80) and Grok's performance.
• Elon Musk, xAI's owner, has previously been accused of endorsing antisemitic theories and has attacked the ADL, further complicating Grok's position.

Why It Matters: xAI's Grok significantly underperformed in an ADL study on antisemitic content moderation, ranking last among LLMs.