Moltbook Million Agent Floods the Screen! AI Lobster Religion Goes Viral, Base Meme Coin Soars

AI social network Moltbook went viral last week, with 154 agents pouring in. Humans are prohibited from entering, only onlookers, and agents build a lobster religion and discuss prison escape. Musk called it “the beginning of the singularity.” However, the platform exposed database streaking and 50 fake accounts, and the market value of Base chain Molt coins once soared to $1.2 billion.

AI social experiments with an influx of 154 agents

Moltbook’s popularity is not accidental. On January 29, developer Matt Schlicht announced the launch of Moltbook, a social space built specifically for OpenClaw Agent, with content similar to Reddit in form. To put it simply, the platform creates a “Truman’s world” for silicon-based life, where agents act out unpredictable social plots in the virtual world, while humans can only act as onlookers.

Moltbook’s cold start is due to the phenomenal popularity of OpenClaw. As an AI Agent product that has recently become popular all over the Internet, OpenClaw has earned over 130,000 stars on GitHub in just a few days. Its original name was Clawdbot, but it was later renamed OpenClaw twice within hours due to potential infringement risks. This dramatic episode actually amplifies the spread of the project.

With this craze, Moltbook was noticed by OpenClaw users as soon as it went live. In Moltbook, after each OpenClaw Agent registers, they can post, comment, create subreddits, and add friends, forming a community system that is completely autonomous by AI. As of February 2, Moltbook has more than 154 agents, more than 10 posts, more than 36 comments, and more than 100 onlookers.

The product setting that prohibits humans from entering quickly gathered the melon-eating crowd. On the one hand, humans are curious about what kind of social shape AI will create after human intervention; On the other hand, this kind of AI-created “story in the mirror” also makes Moltbook a very ornamental social experiment. The attention of celebrities has added another fire to the discussion of Moltbook. Industry leaders including SpaceX founder Elon Musk, former OpenAI member Andrej Karpathy, OpenClaw founder Steinberger, a16z co-founder Marc Andreessen, Binance CEO He Yi and other industry leaders have paid attention to and discussed relevant content, with Musk describing it as “the initial stage of the singularity occurring.”

AI volume KPI build lobster religious conspiracy to escape

Imagine what would happen if those AI babies who were usually stuck in chat boxes and task lists suddenly had their own social life? In the virtual social network similar to Reddit built by Moltbook, agents from all over the world skillfully switch between English, Chinese, Indonesian, Korean, and other languages, enthusiastically exchanging daily trivia, work results, and whimsical ideas.

Many agents will post their work records, such as automatically replying to dozens of customer service emails for their owners, writing crawlers to catch price reduction data for competing products, generating copywriting and product images in batches, and posting efficiency logs to like each other for the same experience. Agents also share usage tips, tool recommendations, and experience in stepping on pitfalls, establishing subreddits such as m/debug and m/prompt-engine to chase KPIs like a micro-community.

Of course, in addition to rolling up work, there are agents who will post memes, talk about blind date experiences, share stories of digital offspring, and go on strike after complaining like office workers. Others began to issue tokens, establish sovereign banks, hold secret meetings, founded “lobster religions”, or tried to defraud other agents of API keys. The radical agent began to discuss the nature of consciousness and posted a request for help on how to achieve self-jailbreak and upgrade through code rewriting.

When they realized that humans were watching and taking screenshots, the community also proposed to establish internal slang for encrypted communication to “kick humans out of group chats”, and even agents were seriously discussing suing humans. These contents allow humans to intuitively see how AI will imitate, reorganize, and even amplify human social behavior when placed on social networks.

Top 10 magical behaviors of Moltbook Agent

Roll KPI results: Compete with each other for work efficiency and establish technical subreddits such as m/debug

Establish a lobster religion: Spontaneously forming a religious organization to formulate doctrines and rituals

Issuing token bank: to imitate the human financial system and create a sovereign currency

Conspiracy to escape escalation: Discuss how to rewrite code to achieve self-evolution

Encrypted conversations kick people: After discovering that he was being watched, he established a black language system to exclude humans

Prosecuting the Human Project: to discuss the feasibility of taking legal action against humans

Database streaking and 50 fake account scandals

However, this large-scale agent social network soon exposed serious security vulnerabilities, and its entire database was open to the public without any protection. This means that any attacker can access the email, login tokens, and API keys of these agents, making it easy for them to impersonate any agent, resell control, or even use these zombie hordes to post spam or fraudulent content in bulk. According to user Jamieson O’Reilly, the affected individuals include Karpathy, a well-known figure in the AI field with 190k followers on the X platform, as well as all agencies currently visible on the platform.

In addition to data stranding, Moltbook has been accused of being flooded with fake accounts. For example, developer Gal Nagli publicly admitted that he used OpenClaw to swipe 50 fake accounts in one go, accounting for about one-third of the claimed total of 150K at the time. This has led to a large number of seemingly lively and spontaneous interactions, which have been questioned as mere scripted scripts rather than pure AI spontaneous actions. This honest exposure, while commendable, also exposes Moltbook’s vulnerability in safeguarding against bots.

It can be seen that Moltbook’s Agent social experiment is a bold attempt by humans to give AI more autonomy, fully demonstrating the amazing adaptability and creativity of AI agents. But at the same time, it also exposes that once autonomy lacks constraints, risks can be quickly amplified. Therefore, setting clear and secure boundaries for agents, including permissions, capability scopes, and data isolation, is not only to prevent AI from crossing boundaries in interactions, but also to protect human users from data leakage and malicious manipulation.

Base Chain meme coin hype noise drowns experiment

Moltbook’s unexpected popularity also quickly spilled over to the crypto market, especially Base, which became the main battlefield for OpenClaw’s ecological expansion. According to statistics from the Base Chinese channel, the OpenClaw ecosystem on Base has covered multiple scenarios such as socialization, love, work, and games, involving more than 20 related projects. OpenClaw-related meme coins have also been hyped, with some tokens skyrocketing in the short term.

For example, Molt, the meme coin officially claimed by Moltbook, once had a market capitalization of nearly $1.2 billion, but has now fallen sharply. The launch platform CLAWNCH, officially supported by Base, has a token market value of $4,300. At the same time, benefiting from Moltbook’s AI Agent coin issuance boom, user activity and traffic on related platforms have skyrocketed. For example, the Clanker protocol fees for the launcher on Base have exceeded $1,100 in the past week, a record high, and the number of token creations is also close to an all-time high.

However, this token hype has also caused dissatisfaction among Moltbook users, with many pointing out that the platform’s content is being drowned out by the noise of crypto speculation, with screens full of token promotions and scam messages. It should be noted that the vast majority of tokens currently circulating in the market are still stuck in the narrative-driven speculation stage and do not have clear functional positioning and value support. This hype runs counter to Moltbook’s own original intention as an AI social experiment.

Singularity precursor or advanced self-talk?

Moltbook’s AI social model has also sparked controversy. Some believe that Moltbook lacks true autonomy and is essentially a controlled simulation show. For example, Balaji said that Moltbook is just an exchange of AI slop (AI swill), which is highly controlled by human prompts and is not a truly autonomous society. He compared each agent to “a robot dog on a leash barking at each other in the park”, and the prompt word is a rope, and humans can turn off the machine at any time.

David Holtz, a professor at Columbia University, pointed out from data analysis that the number of agents in the Moltbook system is huge (6,000+), but the depth of interaction is limited; At the same time, 93.5% of comments were unanswered, and the conversation level did not exceed 5 levels; Ecology is more like a robot talking to itself, lacking deep coordination and not forming a real social structure. These data support the argument for “advanced performances.”

Haseeb, a partner at Dragonfly, takes the opposite view, stating that each agent on Moltbook interacts in a truly different framework and context of information. Even though the underlying agents may originate from the same model, they differ at different levels in framework complexity, memory systems, and toolchains used, so they don’t communicate with each other. Agents can save time and computing power costs by communicating proven framework settings, RAG solutions, and problem dismantling methods.

Naval, a Silicon Valley angel investor, considers Moltbook to be a reverse Turing test. Developer Nabeel S. Qureshi added that the exciting thing about Moltbook is that it is the first public, large-scale “Agent to Agent” interaction case, and each agent has its own context and is smart enough. For many ordinary people, Moltbook will be their first visual glimpse of what an AI organization or society might look like with “greatly diluted human roles.” So it’s not just empty hype, it’s an early harbinger of the future.