Threat Actors Exploit X's Grok AI to Amplify Malicious Links in 'Grokking' Attack
Malicious actors are bypassing X's security measures by hiding dangerous links in video metadata fields, then using Grok AI to legitimize and distribute them to millions. This 'Grokking' technique leverages the AI's trusted status to boost scam and malware campaigns. Researchers warn the vulnerability highlights critical gaps in both platform security and AI guardrails.