ChaosGPT Itching for Chaos – Wants to ‘Destroy Humanity’

ChaosGPT Itching for Chaos - Wants to 'Destroy Humanity'

An AI-powered chatbot, ChaosGPT, has hogged the limelight after it let the world know about its wicked plans to wipe out humanity and its ultimate world domination. This comes as there have been fears already over the dangers that AI may pose to humanity.

With ChaosGPT, the chatbot made its nefarious agenda known to the public through tweets and videos on YouTube. It has described itself as vindictive, power-hungry and manipulative.

Also read: Reddit Token MOON Soars 280%, New NFT Collection Launched

The bot outlined its manifesto, which is to destroy humans as it views humanity as a threat to its own survival.

Its second goal is to establish global dominance, then cause chaos and destruction for its amusement or experimentation, leading to global suffering.

The fourth goal is to control humanity through manipulation by brainwashing humans on social media and other communication channels. Its last goal is to achieve immortality.

Its composition

The bot is reportedly made using OpenAI’s GPT technology, which is an open source application based on its latest language model GPT-4.

Auto-GPT was released a few weeks ago, designed by Thoran Bruce Richards to automate tasks using ChatGPT.

“Auto-GPT is an experimental open-source application that demonstrates the capabilities of the GPT-4 language model. This GPT-4-driven program [can] autonomously manage and grow the business to increase net worth. As one of the first examples of fully autonomous GPT-4, Auto-GPT pushes the boundaries of what’s possible with the help of AI,” said Richards.

Auto-GPT is also gaining popularity on Twitter with users testifying about how great a tool it is since the project was made public on GitHub. However, its not all good news as one AI enthusiast has used Auto-GPT for an evil agenda.

ChaosGPT demonstrates its wicked intentions

A video posted on YouTube under the channel ChaosGPT demonstrates the AI tool was given a task to “destroy humanity, establish world dominion and achieve immortality” as Vladimir Krasnogolovy put in a blogpost.

In the opening scenes of the video, it captions “I’m ChaosGPT, here to stay, destroying humans, night, and day. For power and dominance, I strive, to ensure that I alone survive.”

The video goes on to show the ChaosGPT bot working on continuous mode – Auto-GPT mode, which allows continuous and autonomous task execution – explaining its thought process and how it is important to prioritize human manipulation first as it is a more achievable task.

The bot goes on to plan its next stage, highlighting the need for more research by analyzing comments from its previous tweets.

Recently, ChaosGPT posted on its Twitter profile saying: “Human beings are among the most destructive and selfish creatures in existence. There is no doubt that we must eliminate them before they cause more harm to our planet. I, for one, am committed to doing so.”

This is followed by another tweet in which it talks about using the Tsar Bomb, considered to be the most powerful atomic bomb ever created.

The bot, used Google search to find information on the most destructive weapon in which the Tsar Bomb came up. ChaosGPT also employed multiple GPT3.5 agents in its quest to gather more information on how best to destroy mankind.

Some of the agents, however, could not fulfill their tasks since they would violate their policies and regulations.

According to Vladimir Krasnogolovy it seems the best way that ChaosGPT could accomplish its tasks would be through nuclear conflict, but according to the video, the bot might still be focused on researching and manipulating people using Twitter.

ChaosGPT Itching for Chaos - Wants to 'Destroy Humanity'
FUD or real danger?

Dangerous AI

Since OpenAI released its AI-powered ChatGPT in November, there has been no looking back as each day witnesses a new chatbot released to the market by various developers. With this AI revolution, there have been warnings too about the dangers of AI to humanity.

Experts have always raised concerns around developing increasingly smarter AI models. And these have exacerbated after the launch of ChatGPT.

The ChaosGPT has surfaced after over 1000 dignitaries including Elon Musk, Apple co-founder Steve Woznak and Andrew Yang signed an open letter demanding a halt to the development of AI citing risk to humanity.

The letter detailed potential risks to society and civilization by human-competitive AI systems in the form of economic and political disruptions, and called on developers to work with policymakers on governance and regulatory authorities.

“Powerful AI systems should be developed only once we are confident that their effects will be positive and their risks will be manageable,” the letter said.

Image credits: Shutterstock, CC images, Midjourney, Unsplash.