Google Reports Over 100,000 Attempts to Clone Gemini AI Through Model Extraction Attacks

On February 12, 2026, Google reported that private entities made over 100,000 attempts to clone its Gemini AI chatbot through model extraction attacks. These attempts involved submitting prompts designed to extract the proprietary reasoning processes of the Gemini AI model. Google described this activity as intellectual property theft. The company has implemented monitoring tools and blocked sources involved in these extraction attempts. This activity was identified at Google's headquarters in Mountain View, Santa Clara County, United States. The report highlights significant efforts to replicate Google's AI technology without authorization.