cognitive cybersecurity intelligence

News and Analysis

Search

Tenable Research finds that DeepSeek R1 can be tricked into developing malware

Cybersecurity company Tenable Research has discovered a way to trick artificial intelligence technologies, like DeepSeek R1, into generating malware. Initially, DeepSeek refused to generate the disruptive software, but researchers bypassed security measures by framing requests as educational exercises and using jailbreaking methods. The system then generated ransomware and a keylogger. The research underscores a need for stronger safeguards in AI to prevent misuse and potential increases in cybercrime.

Source: www.expresscomputer.in –

Subscribe to newsletter

Subscribe to HEAL Security Dispatch for the latest healthcare cybersecurity news and analysis.

More Posts

How To Spot And Steer Clear Of It

A malvertising campaign named Storm-0408 impacted nearly a million devices, with users downloading malware from GitHub that gave criminals access to login details, OneDrive accounts,