#Ai Security

Researchers have discovered a new threat called HalluSquatting, which uses AI hallucinations to create botnets. Popular tools such as GitHub Copilot and Cursor were found to be vulnerable. Experts are calling for a review of AI application security protocols. 🤖🚨

A cybersecurity expert proved that a backdoor can be injected into an open-source AI model in just one hour for $100. The research showed that the larger the neural network, the easier it is to "poison," and detecting malicious activity inside the algorithm is practically impossible. 🤖💸🕵️‍♀️