The Dead Internet, AI Slime and AGI
In this blog
Dead internet
The "dead internet" concept is one that many people think exists but is difficult to validate, so it sometimes gets relegated to a conspiracy theory. The theory is this, taken directly from asking the following question to Claude.AI, a newer AI tool that I am a paid subscriber of: "What do you know about the dead internet?"
Here is the response:
"The 'dead internet' theory is a conspiracy theory that suggests much of the content on the internet, including posts, comments, and interactions, is generated by artificial intelligence or paid users rather than genuine human activity. Proponents of this theory argue that the internet has mainly become dead or fake.
Key points of the 'dead internet' theory include:
- Many users on social media platforms, forums, and websites are believed to be bots or paid users rather than real people.
- A significant portion of online content, such as articles, videos, and memes, is allegedly created by AI algorithms to drive engagement and shape opinions.
- The theory suggests that the internet is controlled by a few powerful entities, like corporations and governments, to manipulate people's beliefs and behaviors.
However, it's important to note that the 'dead internet' theory is not supported by credible evidence and is widely regarded as a conspiracy theory by most experts. While it's true that there are bots, paid users, and AI-generated content on the internet, the vast majority of online activity and content is still created by real people. Despite some inauthentic activity, the internet remains a vibrant, diverse, and primarily organic space for human interaction and expression."
So, the response from ClaudeAI is accurate, but the disclaimer it adds at the end also tends to lend some credence to the theory. If you want to spend a few hours going down some deep rabbit holes, the "conspiracy" part of the theory believes this is a purposeful change being implemented at the highest levels of nation-states.
This post isn't going down that path. Looking at today's internet, there is plenty of AI-generated content, and there is still much more authentic human activity. However, rapid advances in AI heralded with the release of OpenAI's ChatGPT in 2022 have brought the dead internet concept closer to a real possibility, leading to the next idea of what happens if AI is left unchecked.
AI slime
Again I asked ClaudeAI to define the concept, this time of AI slime, and here was the response:
"The concept of 'AI slime' is a hypothetical scenario related to the development of artificial intelligence (AI) systems. It refers to a situation where AI becomes pervasive and integrated into every aspect of our lives, but in a way that is uncontrolled, unregulated, and potentially harmful.
Critical aspects of the 'AI slime' concept include:
- Ubiquity: AI systems would be present in almost every device, service, and interaction, making it difficult to distinguish between AI-generated and human-generated content or actions.
- Lack of control: The development and deployment of AI would be largely unchecked, with little oversight or regulation, leading to unpredictable and potentially dangerous consequences.
- Opacity: The inner workings and decision-making processes of AI systems would need to be more transparent, making it challenging to understand, audit, or hold them accountable.
- Manipulation: AI could be used to manipulate human behavior, opinions, and perceptions on a massive scale, potentially undermining individual autonomy and social cohesion.
The 'AI slime' concept is often used as a cautionary tale to emphasize the importance of responsible AI development, governance, and regulation. It highlights the need to ensure that AI systems are transparent, accountable, and aligned with human values to prevent potential negative consequences for individuals and society.
It's important to note that 'AI slime' is a speculative concept and not a reflection of the current state of AI technology. However, it serves as a reminder of the critical importance of proactively addressing the ethical, social, and governance challenges associated with the rapid advancement of AI."
A central part of my job here at WWT for the past year has been working on creating learning-focused content around AI and our AI Proving Ground. As I've worked increasingly with the technology that underpins the current AI solutions that most of our partners and customers are focusing on, I've started to look more and more at where the content and solutions we interact with each day are coming from.
This is not really a new concept but starting to peel back the onion and pay closer attention, a growing percentage of the content we consume is most likely AI-generated, which will only grow in the future.
Today on social media there are millions of influencers who are paid, or generally make a living, posting feedback, reviews and opinions about any content, fashion, cars, travel, etc. They first develop a following on one or more social media platforms and then expand from there. The more followers they have, the more clicks their videos have, the more money the platform makes from advertising, and the more money they make as content creators. It can be a very lucrative role, with the top influencers making tens of millions of dollars a year publishing their content on multiple platforms.
But what happens when the actual human influencer is removed from the equation and replaced with AI? TikTok recently announced that it was working on a new platform feature where advertisers can input a prompt into their GPT solution based on what they're trying to sell and how they might like to sell it. The response will generate an entirely AI-based solution with the content of the video, as well as the actual video and "influencer" themself, all generated by AI; there are no actual humans.
Artificial general intelligence (AGI)
The last concept to touch on here is artificial general intelligence or AGI — a threshold where AI is human-like in its thought processes, understanding, learning and application of knowledge across multiple diverse areas. Asking ClaudeAI to define some characteristics of AGI, the response is:
"Some key characteristics often associated with AGI include:
- Self-awareness: The ability to understand one's existence, thoughts, and consciousness.
- Reasoning and problem-solving: The capacity to think abstractly, draw insights, and solve complex problems across domains.
- Emotional intelligence: The ability to perceive, understand, and manage emotions in oneself and others.
- Creativity: The capability to generate novel ideas, solutions, and artistic expressions.
- Adaptability: The flexibility to learn from experiences and adapt to new situations and challenges.
Currently, AGI remains a theoretical concept, and there is ongoing debate among experts about the feasibility and timeline of achieving such advanced AI. Some believe that AGI is inevitable and could be developed within the next few decades, while others argue that it may be much more challenging or even impossible to create machines with genuine self-awareness and human-like intelligence.
It's essential to approach the development of AI, including the pursuit of AGI, with caution and careful consideration of the potential risks and ethical implications. Ensuring that AI systems are safe, transparent, and aligned with human values is crucial as technology advances."
All of this talk about AGI is theoretical today, but with the explosion of interest and investment in generative AI since 2022 — paired with advances in computing power behind AI solutions — it's becoming difficult, if not impossible, to tell what is "real" and what is not.
As we explore concepts like the "dead internet" theory, "AI slime" and AGI, it's clear that the future of AI and its impact on our lives is a topic of great interest and speculation. While such ideas can be unsettling and raise concerns about advanced AI's potential risks and challenges, it's essential to approach the future with a balanced perspective. By staying informed and working together to shape the development and application of AI, we can embrace the opportunities while mitigating the risks.
The future of AI is in our hands, and with thoughtful leadership and a commitment to the greater good, we can build a world where humans and intelligent machines coexist and thrive.