Bots Will Dominate Internet by 2027

The Rise of the Machines: How Bots Are Taking Over the Internet

In the not-so-distant future, you might find yourself having more meaningful conversations with algorithms than with actual humans online. According to a growing body of evidence, the internet as we know it—once a bustling digital agora of human creativity and connection—could be more “dead” than alive within just three years. This isn’t science fiction; it’s what cybersecurity researchers and internet culture experts are calling the inevitable result of our AI-fueled digital arms race.

What Is the “Dead Internet Theory”?

First surfacing in 2021 as a fringe idea on internet forums, the “dead internet theory” has rapidly gained traction among digital natives and tech critics alike. The theory posits that the internet is transitioning from a human-driven ecosystem to one dominated by autonomous bots and algorithmic content generation. What started as an obscure corner of internet culture has now become a mainstream concern, fueled by undeniable statistics and increasingly obvious signs in our daily online interactions.

As Popular Mechanics science editor Darren Orf explains, the internet is becoming “a future internet where bot-driven interactions far outnumber human ones.” This shift isn’t just theoretical—it’s measurable and accelerating.

The Numbers Don’t Lie

Cybersecurity firm Imperva’s 2024 “Bad Bot” report provides chilling evidence of this trend. In 2021, bots accounted for 42.3 percent of all internet traffic. By 2023, that number had jumped to 49.6 percent—a dramatic increase that suggests bots are not just keeping pace with human users, but outpacing them. The 2025 Imperva Bad Bot Report reveals an even more startling development: bots now generate more than half of global internet traffic (51 percent).

But it gets worse. A large-scale study by Ahrefs revealed that 74.2 percent of newly created web pages in April 2025 included AI-generated content. This suggests that not only are bots consuming more of our internet traffic, they’re also producing an ever-increasing share of new content. Some researchers even suggest that up to 90 percent of online content could be AI-generated by 2025.

The phenomenon isn’t just about quantity—it’s also about quality and authenticity. The Pew Research Center estimates that 38 percent of human-made webpages from 2013 no longer exist, a process known as “link rot.” As human-generated content disappears and bot-generated content proliferates, we’re witnessing what amounts to a fundamental shift in the internet’s ecosystem.

Visualization of bot traffic overtaking human traffic on the internet

Bot traffic has now surpassed human traffic on the internet. Source: Imperva 2025 Bad Bot Report

When Shrimp Jesus Became a Harbinger of Doom

In early 2024, the internet was briefly captivated by an unusual trend: hyper-realistic AI-generated images of shrimp bearing the face of Jesus. What seemed like a harmless internet oddity was actually a revealing glimpse into our bot-dominated future. These “shrimp Jesus” images garnered tens of thousands of likes and comments, demonstrating how easily bot-generated content can capture human attention and engagement.

According to Jake Renzella and Vlada Rozova in The Conversation, this phenomenon represents “something more sinister: It was a subtle glimpse into the emergence of a ‘dead internet.’” Their analysis suggests that what we’re seeing isn’t just quirky AI art—it’s the economic reality of engagement farming, where bots are optimized to extract maximum attention (and thus advertising revenue) from human users.

Economic Incentives Drive the Bot Revolution

The reason for this shift isn’t hard to understand when you consider how the internet economy works. In a system where attention equals dollars via ad revenue, automating the process of capturing that attention becomes an easy way to “print money.” This isn’t conspiracy—it’s capitalism at work, but with algorithms instead of humans pulling the levers.

Taylor Lorenz, a former New York Times and Washington Post reporter who specializes in internet culture, puts it bluntly: “I think the internet was terminally ill before ChatGPT was announced and released.” According to Lorenz, algorithmic ranking systems—themselves a form of artificial intelligence—set the stage for “endless, worthless pieces of content and for the whole internet to be optimized in the most absurd ways.”

When Bots Go Bad: Misinformation and Manipulation

While some bots are benign and even helpful (like search engine crawlers that help organize the web), the more concerning trend involves malicious bots and automated disinformation campaigns. Data analysis company NewGuard’s May 2025 review found over 1,000 news sites run almost entirely by bots, with 167 of those masquerading as legitimate Russian local news websites. These sites publish “egregiously misleading claims about the Ukraine war and primarily use AI to generate content.”

This weaponization of AI content generation represents one of the most significant dangers of a bot-dominated internet. When bots can create convincing fake news articles, social media posts, and even entire websites at scale, the very foundation of online information becomes questionable.

According to Fastly’s Q1 2025 Threat Insights Report, 37% of global internet activity is now automated bot traffic, with a staggering 89% of that traffic deemed unwanted. This suggests that not only are bots taking over the internet, but the majority of their activity is considered harmful by cybersecurity experts.

Chart showing AI crawler traffic by purpose and industry

AI crawlers now make up a significant portion of internet traffic. Source: Cloudflare Radar

The OpenAI Paradox

Perhaps most ironically, some of the companies that created the tools enabling this bot revolution seem surprised by the outcome. When OpenAI CEO Sam Altman was confronted with the prevalence of LLM-run Twitter accounts, he responded that he “never really took dead internet theory seriously, but that there are ‘a lot of LLM-run twitter accounts now,’” seemingly unaware that his own company is one of the chief architects of the modern internet’s decaying state.

What This Means for You and Me

Despite these alarming trends, it’s important to note that this doesn’t mean personal use of the internet will disappear. Your direct messages to friends, private social media posts, and other forms of genuine human communication will likely persist. However, as bot traffic becomes increasingly dominant in public spaces, users will need to become more vigilant in discerning what’s real from what’s artificially generated.

Renzella and Rozova capture the sentiment of many who mourn this transition: “The freedom to create and share our thoughts on the internet and social media is what made it so powerful. This is the sense in which the internet we knew and loved is ‘dead.’” It’s not that the internet is technically dead—it still functions, still connects people—but its character has fundamentally changed.

Looking Ahead: A Bot-Filled Future

Whether the “dead internet theory” fully materializes in the next three years remains to be seen, but the trends are undeniable. As AI technology continues to advance and economic incentives drive further automation, we may find ourselves in a digital landscape where distinguishing between human and machine-generated content becomes increasingly difficult.

The solution isn’t necessarily to halt technological progress, but rather to develop better tools for content verification, transparency in AI-generated content labeling, and perhaps a fundamental rethinking of how we measure and value online engagement. Until then, the internet—at least the public-facing portion of it—may indeed be more dead than alive.