Contractors working on Google Gemini are comparing its responses to Claude's, according to internal correspondence seen by TechCrunch.

Google is reportedly using Anthropic's AI system, Claude, to assess its own model, Gemini. Contractors compare outputs based on criteria like truthfulness and clarity, identifying areas for improvement.

Google is reportedly using Anthropic’s AI model, Claude, to evaluate and enhance the performance of its in-house AI system, Gemini. This strategy involves contractors comparing the responses of both AI models to identical prompts,

Contractors observed that Claude’s safety measures were stricter than Gemini’s, with the former declining to respond to prompts it deemed unsafe.

Google has responded to claims it used Anthropic's Claude AI to evaluate Gemini AI, denying any training involvement.