In a fresh reminder of rising risks around artificial intelligence, attackers have recently attempted to recreate Google’s Gemini AI by aggressively probing the system with massive volumes of queries. The effort highlights how advanced AI models are becoming targets for both cybercriminals and state-backed groups.
Hackers reportedly tried to build their own version of Gemini by flooding it with more than 100,000 questions across multiple languages. The approach relied on a technique known as knowledge distillation. Through this method, attackers study how an AI responds and processes information, allowing them to create a copycat model without accessing or stealing the original source code.
Beyond model replication attempts, some state-sponsored groups went a step further by using Gemini directly during cyber operations. China’s APT31 reportedly used Gemini-powered tools to identify system vulnerabilities in US organisations. Iran’s APT42 used the AI to help craft phishing emails, while North Korea’s UNC2970 analysed employees working in the defence sector with support from the system.
In response, Google has taken action by blocking suspicious accounts and closely monitoring abnormal behaviour linked to attempts at extracting the model. The company has also increased its focus on detecting patterns that suggest automated or coordinated misuse of Gemini.
Notably, Google observed that most of these extraction attempts did not come from government-backed hackers alone. A large number were linked to private companies and independent researchers from different parts of the world, showing how widespread interest in copying advanced AI systems has become.
As techniques for duplicating AI models become easier and less expensive, the challenge of safeguarding systems like Gemini is growing. The latest incidents underline the importance of stronger protections to prevent misuse, abuse, and unauthorised replication of high-value AI technologies.
Also read: Viksit Workforce for a Viksit Bharat
Do Follow: The Mainstream formerly known as CIO News LinkedIn Account | The Mainstream formerly known as CIO News Facebook | The Mainstream formerly known as CIO News Youtube | The Mainstream formerly known as CIO News Twitter
About us:
The Mainstream is a premier platform delivering the latest updates and informed perspectives across the technology business and cyber landscape. Built on research-driven, thought leadership and original intellectual property, The Mainstream also curates summits & conferences that convene decision makers to explore how technology reshapes industries and leadership. With a growing presence in India and globally across the Middle East, Africa, ASEAN, the USA, the UK and Australia, The Mainstream carries a vision to bring the latest happenings and insights to 8.2 billion people and to place technology at the centre of conversation for leaders navigating the future.
