xAI’s Grok 4.5 integrates Muse image generation and offers 40% better truthfulness scores?

xAI’s Grok 4.5 integrates Muse image generation and offers 40% better truthfulness scores

Model supports 256K context tokens and native function calling?

Model supports 256K context tokens and native function calling

Alex Karp says verifiable data is becoming the critical differentiator for enterprise AI adoption

Opinion & Analysis

Stratechery July 10, 2026

⚡xAI drops Grok 4.5 as Palantir’s CEO warns on data verification.

Deep Dive

The battle for verifiable data is increasingly defining the AI race, from Meta to Grok to the frontier labs, according to the article.

Key Points

xAI’s Grok 4.5 integrates Muse image generation and offers 40% better truthfulness scores
Model supports 256K context tokens and native function calling
Alex Karp says verifiable data is becoming the critical differentiator for enterprise AI adoption

Enterprise AI deployments will now prioritize data provenance over raw performance, reshaping model selection.