Amy Fraser4 June 2024 | 12:12

The rise of 'shrimp Jesus' highlights dangers of the AI-driven web

This bizarre phenomenon finds its explanation in the "dead internet theory," which suggests that AI and bot-generated content have outpaced human-generated content on the web.

The rise of 'shrimp Jesus' highlights dangers of the AI-driven web

The rise of 'Shrimp Jesus': The dangers of the AI driven web / X: @natejhake

Lester Kiewit interviews Vlada Rozova, Research Fellow in Applied Machine Learning at the University of Melbourne.

Listen below.

If you were to type 'shrimp Jesus' into social media search bars, you'd likely come across numerous images where artificial intelligence (AI) has merged crustaceans with the iconic image of Jesus Christ. 

This bizarre phenomenon finds its explanation in the "dead internet theory," which suggests that AI and bot-generated content have outpaced human-generated content on the web.

According to this theory, the majority of online activity, including social media posts, is now orchestrated by AI agents rather than humans.

"I think what's happening now is that this conspiracy theory is becoming a little bit of reality, and that's the scary part."
- Vlada Rozova, Research Fellow in Applied Machine Learning – University of Melbourne

While the 'shrimp Jesus' images might initially seem harmless, they hint at a deeper agenda.

As AI-driven accounts amass followers, their seemingly high legitimacy attracts real users, forming a vast network of automated accounts.

Vlada Rozova and Jake Renzella elaborate on this concept in their article "The ‘dead internet theory’ makes eerie claims about an AI-run web. The truth is more sinister," cautioning readers to approach social media and other online platforms with a healthy dose of scepticism.

The dead internet theory serves as a stark reminder to critically evaluate the content we encounter online.

RELATED: ‘Sexualised’ and ‘homoerotic’ Jesus painting sparks outrage in Spain

"AI and bot-generated content has basically surpassed that of human beings."
- Vlada Rozova, Research Fellow in Applied Machine Learning – University of Melbourne

Scroll up to the audio player to listen to the interview.