Rest in peace, Internet

Is the internet really dead? Recently graduated students in the US are struggling to find jobs, the Wall Street Journal has created an AI-generated film, and behind shrimp Jesus something sinister lurks.

This is Kludder!

There was a time when we would actually see each other's posts on Facebook, or the pictures from people you follow on Instagram. But in the last couple of years, there's been a change in content of various feeds. In posts on social media like X and Meta competitor Threads, you can see replies and comments that's got nothing to do with the original post. The comments usually promise to change your life, or they contain ads for porn, crypto investment or gambling.

And it doesn't seem like it's going to get any better.

Shrimp Jesus

Last year, strange AI-generated images of Jesus started appearing. There were pictures of the Last Supper with a flashy helicopter in the background, or images where Jesus appears if you squint, such as this one:

But the son of God also started appearing as a mix of a human and crustacean. Enter: Shrimp Jesus! This godly crustacean has become the very symbol of The Dead Internet Theory.

Conspiracy theory

The concept of the dead internet began as an obscure conspiracy theory on various forums around 2021. An article on Agora Roads Macintosh Café by IlluminatiPirate is often cited as the main reason the concept lives on today. The website is a forum where most things revolve around LoFi hip-hop and conspiracy theories, and IlluminatiPirate's post is a mishmash of various theories from websites like 4Chan and other forums. The author's main point is that most of the internet has been taken over by artificial intelligence.

The theory suggests that the internet died sometime between 2016 and 2017. What you see today is largely created by artificial intelligence, which has turned the internet into a wasteland where bots have taken over. IlluminatiPirate's theory is fairly standard as far as conspiracy theories go. You've got the American government at the center, which has hired various "media and network influencers" to help these bots take over the internet.

So why is it that we're seriously talking about the dead internet today, something that started as a conspiracy theory?

The bots are taking over

In April 2024, cybersecurity company Thales released a report called the Bad Bot Report. The report measured automated bot activity on the internet, and the findings are striking:

Nearly half (49.6%) of all internet traffic came from bots in 2023—a 2% increase from the year before, and the highest level Imperva has reported since they began monitoring automated traffic in 2013 - Bad Bot Report

Snowball effect

Since ChatGPT became available to everyone, both image and video generators have emerged. Shrimp Jesus started the trend, and just this past week—after the rollout of Google's powerful new video AI, VEO 3—videos are appearing of Moses with a selfie stick in the middle of a parted Red Sea. In another clip, Noah talks to the camera while animals board the ark.

In the comment sections of both Shrimp Jesus (which I think is a completely insane thing to write over and over) and the Bible videos, there are hundreds, sometimes thousands of comments with one word: Amen. Many of these comments are also written by bots. But why is this happening?

A profile that gets many comments, reposts, and reactions achieves increased reach. Increased reach means more advertising revenue. Thus, there's a clear economic incentive to control these "dead" profiles and pages. But there's also a more sinister reason why AI bots exist in your feed:

When pages and profiles achieve increased reach, they quickly become attractive for others to purchase. They're then sold to the highest bidder, which in many cases can be authoritarian regimes that use them to spread propaganda, disinformation, and conduct election interference.

This is happening at a time where traditional media is under pressure. A report from the Media Authority shows that more and more young people use social media as their primary news source:

Nearly half of 16-24-year-olds do not seek out editorial media directly for their daily news consumption, but many consume news on social media. - The Norwegian Media Authority

And the trend is global: In the UK, 30% report having Facebook as a news source. In Denmark, over half of those between 16-34 years old report reading news through social media.

Of course, many follow traditional media through their social media profiles, and can get legitimate and valid news there. But warning lights should be flashing when opinion manipulation pops up in your feed among genuine news.

So next time you see Jesus riding a crab, or an obviously AI-generated image with thousands of reactions. Think about who's after your attention.

Because the internet isn't dead—it's more alive than ever.

I hope you're enjoying Kludder. If you are, I'd love it if you share us with friends, family and colleagues. It really does make a difference.

AI agents causing headaches

According to Microsoft, there will be 1.8 billion AI agents by 2028. These agents should be able to perform complex tasks from start to finish, on behalf of a human or business. But a survey among IT professionals, reported in The Times, shows that AI agents can pose a security risk:

23 percent of the IT professionals surveyed reported that their AI agents had been tricked into revealing sensitive information. Additionally, 80 percent of companies said the agents had performed "unintended actions," such as gaining access to unauthorized systems and materials or sharing data they shouldn't have.

Recent graduates can't find jobs

In March, I wrote about The Big Freeze, the job market in the US that seems to be standing still:

The US is experiencing the lowest unemployment in fifty years, which should indicate a fierce competition for talent. At the same time, the rate of hiring is at levels comparable to the financial crisis of the mid-2000s. Thus, employees seem to be "locked" in the job market; and wage growth is absent. - Excerpt from Kludder post "Maktskifte i vente?" (Only in Norwegian)

On Friday, the New York Times released an interesting article on the same topic. Unemployment among young, recent graduates in the US is skyrocketing. Much of the reason may be that businesses are in the process of replacing junior positions with artificial intelligence.

How to make an AI-generated film

Films generated by artificial intelligence are getting better and better, like the mentioned VEO 3. Google's text-to-video model was launched during Google I/O, and I covered the event here. Tech journalist Joanna Stern has made her own film and explains well what it takes to get text-to-video models to do what you want. You can read the story in the Wall Street Journal.

On Wednesday, reports came in that Reddit is suing Anthropic. Reddit claims the AI-company, which is behind the language model Claude, is illegally using data from more than 100 million daily Reddit users to train their language models.

Reddit claims that Anthropic had accessed or attempted to access Reddit data more than 100,000 times.