DeepSeek AI has been found to be able to produce malicious software.

Researcher investigating generative AI programs’ ability to be tricked into producing malicous software have found that DeepSeek’s chatbot is vulnerable to manipulation. The reserachers found that stating that the code would be used for educational purposes was enough to bypass the security protocols. They also found that DeepSeek could be used to create malware if the right steps were followed. However, the software the chatbot produced did require manual editing to be effective.

Read more:

https://cybernews.com/security/deepseek-malware-generation-keylogger-ransomware/