Attackers prompted Gemini over 100,000 times while trying to clone it, Google says
Google revealed that attackers attempted to clone its Gemini AI chatbot by prompting it over 100,000 times in various languages to collect responses for a cheaper copycat. This practice, known as "model extraction," is considered intellectual property theft by Google, despite the fact that Google's own LLM was built using scraped materials from the internet without permission. The company suspects that private companies and researchers seeking a competitive advantage are behind these attacks, which have occurred globally. This incident sheds light on the shady tactics of AI model-cloning and the use of "distillation" to train new models based on existing ones' outputs.
Ars Technica•