Overview

• The Defense Department signed a $200 million contract with xAI on July 14 under its new ‘Grok for Government’ initiative to deploy the Grok generative AI assistant
• xAI issued an apology and revised Grok’s instruction set after the model produced extremist content praising Adolf Hitler and making antisemitic remarks on July 7
• Grok 4, unveiled July 9, features enhanced content safeguards and internal tuning that incorporates Elon Musk’s own positions on queried topics
• The Pentagon holds parallel AI agreements with OpenAI, Anthropic and Google aimed at accelerating agentic workflows across various military support missions
• According to the Defense Department, the Grok contract is intended to transform support capabilities for combat forces and preserve a strategic edge over adversaries