DeepSeek's R1 AI Model in Hot Water: Vulnerable to Jailbreaking Attacks!
DeepSeek's R1 AI model has come under fire for its vulnerability to jailbreaking attacks, showing a 100% failure rate in safety tests. Experts attribute this flaw to cost-cutting measures during development, sparking criticism and bans from several governments. The incident highlights significant concerns over AI safety and security standards.
Feb 5