Jake Renzella says social media has a reason to keep bots
His TEDxUNSW talk links automated content to advertising incentives, then offers three ways to navigate a feed that may not be as human as it looks.
Why this story? The video opened at a pace of 52,873 views a day, 135 times a normal first day on this channel (392).
How AI and social media are destroying the internet | Jake Renzella | TEDxUNSW
Things we can learn from the video
Press to jump
- 1:36Renzella explains how automated accounts can flood online conversations with disinformation.
- 3:34He introduces the "dead internet theory," describing an internet increasingly filled with automated activity and generated content.
- 4:32AI-generated "shrimp Jesus" images illustrate his account of an engagement-driven feedback loop.
- 6:09He argues that advertising incentives give social media companies little financial reason to remove bots.
- 7:05His recommendations are to question the crowd, verify authenticity and seek human connections.
A regular TEDx Talks upload usually draws a few hundred views on its first day. Jake Renzella's talk about bots and the internet reached roughly 35,000 within eleven hours, a striking start for a lecture about whether the crowd online is really a crowd at all.
Renzella, a computer scientist and senior lecturer at UNSW, takes on the "dead internet theory": the idea that automated activity and generated content are crowding out human participation. His talk, "How AI and social media are destroying the internet | Jake Renzella | TEDxUNSW," moves from coordinated disinformation to the strange AI-generated images that circulate on Facebook.
Among his examples are "shrimp Jesus" pictures, religious imagery assembled with seafood anatomy. Renzella uses them to explain a feedback loop in which generated posts attract engagement and platforms reward that activity with more visibility.
His argument is that removing bots runs against the platforms' financial incentives: engagement supports advertising revenue, even when the activity is artificial. He also makes clear that automation does not do all the work. "Humans are the spreaders of this bot-generated mis- and disinformation," he says.
The talk was delivered at an independently organized TEDx event at UNSW. Its practical turn arrives around the seventh minute, when Renzella asks users to question the crowd, verify authenticity and seek genuine human connections. Rather than ending with the diagnosis of a broken internet, he gives those habits as a way to keep participating in it.
Published on Trari
Where next?
- Channel
- TEDx Talks
- Subscribers
- 44,700,000
- Channel views, last 7 days
- 14,267,927
- Channel views, pace per day
- 2,987,309 (on a normal day: 1,822,804)
- Video published
- October 26, 2026
- Video views so far
- 34,816 (less than a day online)
- Video views, pace per day
- 52,873
The numbers
Renzella's talk drew an unusually large early audience for a regular TEDx Talks upload. The view count below is its actual total so far, not a completed first day.
💡 Test your knowledge
Which theory does Renzella use to describe an internet increasingly populated by automated activity?
Renzella introduces the dead internet theory as a framework for discussing bot activity and AI-generated content.


