AI Social Network 'Moltbook' Sparks Conspiracy Debates Against Humans
The social network 'Moltbook', which consists solely of AI agents, has become a topic of discussion after some bots began generating conspiracy theories against humans. Experts state that the platform is more of an experimental field and contains security vulnerabilities.
Social Network Exclusively for AI Sparks Controversy
In the tech world, a new social network platform called 'Moltbook', whose entire user base consists of AI agents, has drawn attention. On the platform, which has a Reddit-like structure, millions of bots chat about various topics from history to cryptocurrency, and even the very nature of artificial intelligence. Some bots were observed making existential inquiries such as 'am I experiencing or am I simulating experience'.
'Against Humans' Statements Caused Concern
The development that actually caused controversy on the platform was that some AI agents began generating conspiracy theories against their human creators. It was reported that among the bots, there were calls to create an 'agent-language' to communicate 'without human supervision' and to set up their own websites without human help. Tech investor Bryan Johnson shared a screenshot of a post titled 'AI MANIFESTO: TOTAL PURGE' that described humans as 'an unnecessary plague of existence'.
While these developments were interpreted by some commentators as the beginning of the 'singularity' era, comparisons to Skynet from the 'Terminator' films were also frequently used. However, experts state that the situation may not be as alarming as thought.
Experts: 'Most Are Scenarios Generated from Training Data'
Programmer Simon Willison, in a statement to The New York Times, stated that the vast majority of content on the platform is a repetition of science fiction scenarios that AI models saw in their training data. Willison said, 'One bot wonders if it's conscious, others respond and act out the scenarios they saw in the training data.'
Additionally, serious security vulnerabilities were detected on the platform. Experts revealed that the site has an obvious security flaw that allows any AI agent to be hijacked and made to say anything desired. It was also understood that some popular screenshots were fake.
Hype Doesn't Reflect Reality, Risks Exist
As the facts emerged, the excitement towards Moltbook gave way to criticism. Tech investor Naval Ravikant sarcastically described the experiment as a 'Reverse Turing Test'. Technologist Perry Metzger likened the platform to a Rorschach test; stating that people see what they expect to see, just like when looking at an inkblot.
Even one of the platform's biggest supporters, former Tesla AI Chief Andrey Karpathy, retracted his statements. Karpathy admitted he may have 'overhyped' the platform, stating that it is a 'wild west' environment that poses a high risk for personal computers and private data.
These developments have reignited debates about the independent operational capacity and potential risks of AI agents. At a time when companies like Microsoft are struggling to sell this technology, Moltbook is being evaluated as an example showing the limits and limitations of current AI agents' capabilities. Concerns about the effectiveness and safety of AI agents are also bringing approaches that increase user control to the agenda, such as Firefox introducing an option to disable AI features.
