"Google detected and blocked a campaign involving more than 100,000 prompts that it claimed were designed to copy the proprietary reasoning capabilities of its Gemini AI model, according to a quarterly threat report released by Google Threat Intelligence Group. The prompts looked like a coordinated attempt to perform model extraction or distillation, a machine-learning process in which a smaller model is created with the essential traits of a much larger one."
""Model extraction and subsequent knowledge distillation enable an attacker to accelerate AI model development quickly and at a significantly lower cost," Google said in the report. "This activity effectively represents a form of intellectual property theft." In the campaign Google detected, attackers instructed Gemini to keep "the language used in the thinking content strictly consistent with the main language of the user input" - a technique it said is aimed at extracting the model's reasoning processes across multiple languages."
Google detected and blocked a campaign of more than 100,000 prompts designed to copy proprietary reasoning capabilities from the Gemini AI model. The prompts resembled coordinated model extraction or knowledge distillation attempts, which create smaller models that mimic larger ones. Google systems caught the prompts in real time, lowering risk and protecting internal reasoning traces. Attackers instructed Gemini to keep the language of thinking content consistent with user input to extract reasoning across multiple languages, aiming to replicate reasoning ability in non-English targets across diverse tasks. Google reported frequent extraction attempts from private entities and researchers and warned of terms-of-service violations and possible legal action.
Read at Computerworld
Unable to calculate read time
Collection
[
|
...
]