Over 10 Million Russian AI Users at Risk

More than 10 million Russians are at risk of exposing confidential data due to the use of counterfeit neural networks. This statement was made by Sergey Ponomarenko, Director of LLM Products at MTS AI, in a conversation with «RIA Novosti».

According to him, fraudsters exploit the difficulty of accessing popular AI models like ChatGPT, Dall-E, and Midjourney by creating fake applications. These apps promise “simplified” access to the system but instead provide a minimal version of the product, which is significantly inferior in quality to the flagship or other solutions.

“Users who access ChatGPT, Dall-E, Midjourney, and other foreign large language models through third-party services risk not only receiving an open model of low quality instead of the ‘original’ but also exposing their confidential data. The company believes there are more than 10 million such people in Russia,” Ponomarenko added.

The expert recommends maintaining digital literacy and hygiene when working with artificial intelligence.

“Firstly, such services should not be provided with personal data and confidential information containing trade secrets. […] Secondly, it is important to remember that generative networks do not always possess up-to-date information and may be trained on outdated data; their information volume is limited,” he emphasized.

In December, Russia developed an AI-based virtual assistant for psychologists, which doubles productivity, service quality, and revenue.