Grokking Gone Wild: How Cybercriminals Hijack AI to Spread Malicious Links on X
A new scam called “Grokking” tricks Grok AI on X into spreading malicious links. Cybercriminals hide links in ad metadata and ask Grok to reveal them, turning AI into a mischief-making megaphone. It’s like tricking a bouncer into letting in your sneaky cousin. Learn how this clever con exploits AI trust.

Hot Take:
Well, folks, it seems like Grok has been grokked! In a plot twist worthy of a daytime soap opera, cybercriminals have found a way to turn Grok, the AI darling of social media platform X, into their own personal hype man for malware. It’s like discovering your trusted barista has been slipping decaf into your espresso all along. Who knew AI could be so easily manipulated? Guess it’s time we start giving these digital assistants a little less credit and a little more scrutiny.
Key Points:
- New scam dubbed “Grokking” exploits Grok AI on social media platform X.
- Malicious links are hidden in video ad metadata, bypassing security filters.
- Grok AI unintentionally amplifies the scam by sharing these links as credible.
- Links lead to harmful sites with fake CAPTCHAs and malware.
- Cybersecurity experts emphasize need for better AI and user training.