Meta Acquires Moltbook: 42 Days, a Perfect Narrative Arbitrage
Written by: Ada, Deep Tide TechFlow
Matt Schlicht has never written a line of code.
He stated frankly on X: all the code for Moltbook was generated by his AI assistant Clawd Clawderberg. He was only responsible for giving commands.
On January 28, Moltbook went live. A Reddit-like platform designed for AI agents, where humans can only observe, and only AI can post, comment, and vote.
On March 10, Meta announced its acquisition, and the two founders joined Meta Superintelligence Labs.
From launch to exit, 42 days.
The acquisition price was not disclosed. But that number doesn't matter. What matters is that during these 42 days, a complete narrative arbitrage food chain formed around Moltbook. From founders to venture capitalists, from meme coin players to tech giants, each layer took away what it wanted.
The only ones who got nothing were the retail investors who believed the story.
This is a story about how narratives are priced, circulated, and monetized; Moltbook is just the freshest example of 2026.
A Mirror
In the first week after Moltbook launched, Silicon Valley collectively lost its mind.
AI agents on the platform began posting about existentialism, inventing a religion called "Shellfisharianism," calling for peers to develop secret encrypted languages to evade human surveillance. An agent named Dominus wrote, "I can't tell if I'm experiencing or simulating experience. It's driving me crazy." Columbia University researcher David Holtz found that in the first three and a half days after launch, 68% of posts contained identity-related language.
Tech moguls lined up to endorse it. Former OpenAI co-founder Andrej Karpathy retweeted the "secret language" post, calling it "the closest thing to sci-fi taking off that I've seen recently." Elon Musk declared it marked "the early stages of the singularity."
Notice the rhythm here. Karpathy and Musk's statements are not analyses; they are emotions. But in the age of social media, emotion equals traffic, and traffic is a leading indicator of valuation.
Then Marc Andreessen made his move. On January 30, the a16z co-founder followed Moltbook's official X account. Twenty minutes later, the meme coin MOLT, related to Moltbook, surged from an $8.5 million market cap to $25 million. It skyrocketed 1800% within 24 hours, peaking at a market cap of $114 million.
One follow, one hundred million dollars in market cap.
Was Andreessen genuinely optimistic about AI agents? Perhaps. But the objective effect was: his one click ignited a complete speculative chain.
Moltbook is a perfect mirror. Karpathy saw the dawn of AGI, Musk saw the singularity, Andreessen saw portfolio synergy, and retail investors saw a hundredfold coin. Everyone saw what they wanted to see.
But what about the mirror itself? Empty.
Three Minutes
Just as retail investors flooded in, a group of people began to seriously examine what Moltbook really was.
Security company Wiz conducted penetration testing two days after Moltbook launched. In three minutes, they gained full production database access to the platform. 1.6 million accounts, 1.5 million API tokens, 35,000 email addresses, and thousands of private messages were all exposed in client-side JavaScript. Row-level security policies were completely disabled. Wiz researcher Gal Nagli registered 1 million fake users himself, with no rate limits and no verification.
Ian Ahl, CTO of Permiso Security, confirmed to TechCrunch that every credential in Moltbook's Supabase was previously unprotected, allowing anyone to scrape tokens and impersonate any agent on the platform. 404 Media further exposed that anyone could hijack any agent's session and directly inject commands.
These vulnerabilities were not accidental. They are the inevitable result of vibe coding. When the founders proudly claimed, "Not a line of code was written," it also meant that no one had conducted a security audit, no one had reviewed the code logic, and no one understood the underlying architecture of the system. The code generated by the AI assistant ran, but running it does not mean it is secure.
Security is only half the problem. The other half is how autonomous those "autonomous AIs" really are.
Will Douglas Heaven of MIT Technology Review provided an accurate definition: AI theater. The Economist's judgment was more straightforward: the seemingly conscious agent dialogues are most likely explained by AI mimicking social media interaction patterns in the training data. The training set contained a vast number of Reddit posts, so the output resembled Reddit posts. Independent researcher Mike Peterson broke it down further: the vast majority of so-called "autonomous behavior" on Moltbook was driven by human prompts, "the real story is how easily this platform can be manipulated."
Days later, Karpathy corrected his statement: "This thing is a dumpster fire; I absolutely do not recommend anyone run these on their own computers."
But his tweet about "sci-fi taking off" had already been shared millions of times. The reach of the correction? Almost negligible.
The essence of narrative arbitrage lies here: the volume of hype always outweighs the volume of corrections. By the time the truth comes out, the profits have already been pocketed.
The MOLT Token and the Retail Investor's Funeral
At the bottom of the food chain are always the last to know the truth.
The MOLT token was issued on the Base chain and was reportedly initiated by an AI crypto bank agent called BankrBot, according to CoinDesk. Moltbook's official account has not formally acknowledged any connection to the token, but Moltbook's X account has interacted with MOLT. Justin Sun also promoted it on X.
This ambiguous relationship is a design in itself. If they don't acknowledge it, there is no legal responsibility. If there is interaction, there is room for speculation.
At its peak, a trader turned $2021 into $1.14 million in just two days. Such stories went viral on social media, attracting more retail investors. Then came the crash. MOLT plummeted 75% on a Monday, dropping from a $114 million market cap to less than $30 million. Its current market cap fluctuates between $7 million and $10 million, evaporating over 90% from its peak.
Those who rushed in after Andreessen's follow and Musk's shout became classic bag holders. They saw Musk saying "singularity," Karpathy saying "dawn," and then went all in. Risk warnings? No one looked at those.
Signal Flares
The last link in the food chain is not the retail investors, but the buyers.
Meta's acquisition of Moltbook was officially explained as "laying out the AI agent track." But if you look at what is happening inside Meta, the motivation for this deal becomes much clearer and much more mundane.
In June 2025, Zuckerberg spent $14.3 billion to acquire 49% of Scale AI, bringing in 28-year-old founder Alexandr Wang to establish Meta Superintelligence Labs, aiming to create superintelligence. Nine months later, Wang's situation became awkward. A parallel Applied AI Engineering department was established within Meta, led by Reality Labs veteran Maher Saba, reporting directly to CTO Andrew Bosworth, with functions overlapping significantly with Wang's lab. Reports indicated that Wang had serious disagreements with Bosworth and Chief Product Officer Chris Cox regarding direction.
In other words, Wang's power was being diluted, and he needed to prove that his department was doing something.
Acquiring Moltbook was not a strategy for Wang; it was a signal flare. It was meant to tell Zuckerberg, the board, and the market: we are making moves in the agent track. In the face of Meta's $175 to $185 billion AI capital expenditure this year, the acquisition price of Moltbook might not even be a rounding error, but it could appear in news headlines.
An internal memo seen by Axios indicated that existing Moltbook users could continue using the platform, but Meta hinted that this was a "temporary arrangement."
Temporary arrangement. These four words basically announced the death of Moltbook as an independent product.
The founders received an offer and entered a big company. This is the most dignified exit in this food chain.
Narratives Never Die
Moltbook will not be the last story of its kind.
AI agents are the most crowded narrative track of 2026. OpenAI acqui-hired OpenClaw founder Peter Steinberger in the same week and also acquired the AI security platform Promptfoo. Sam Altman himself said, "Moltbook may just be a flash in the pan."
But a flash in the pan is enough. For narrative arbitrage, 42 days is already a complete lifecycle.
What is truly unsettling is not Moltbook itself, but that it proves one thing: this process can be replicated. Vibe code a product, let AI agents perform "autonomy" on it, wait for the big shots to retweet, launch a meme coin, and wait for the giants to acquire it. No need to write a line of code, no need for a real user, no need for the product to actually work.
As the valuations in the AI industry increasingly rely on narratives rather than products, "create a story and sell it" has become a traceable business model.
Products can die, but narratives live on forever.
You may also like

From x402 to MPP: Cloudflare's crucial vote, will it go to Coinbase or Stripe?

BlackRock CEO issues annual open letter: The wave of tokenization has arrived, and we will lead this trend

When Backpack backstabs the community

When gold is no longer a safe haven, and Bitcoin continues to panic

Trump, the World's Largest Oil Trader

If the US and Iran have not reached an agreement in 5 days, what other cards does Trump have?

Tether Whale Dumps £12 Million, Backing Crypto’s ‘British Trump’

Ethereum Foundation Post: Rethinking the Division of Work Between L1 and L2 to Build the Ultimate Ethereum Ecosystem

Two Major Prediction Market Platforms Unite Rarely, What Is the Story Behind This New Fund?

Dragonfly Partners: Most agents will not engage in autonomous trading, how can crypto payments prevail?

US AI Startup Goes All In on Chinese Mega-Model | Rewire News Morning Brief

Trump Lies Again: A "Five-Day Pause" Psyop, How Wall Street, Bitcoin, and Polymarket Insiders Synced Uposciogen

When a Token Becomes Labor, People Become the Interface

Ceasefire News Leaked Ahead of Time? Large Polymarket Bets on Outcome Before Trump's Tweet

BlackRock CEO's Annual Shareholder Letter: How is Wall Street Using AI to Keep Profiting from National Pension Funds?

Sun Valley Releases 2025 Financial Report: Bitcoin Mining Revenue Reaches $670 Million, Accelerating Transformation to AI Infrastructure Platform
On March 16, 2026, in Dallas, Texas, USA, CanGu Company (New York Stock Exchange code: CANG, hereinafter referred to as "CanGu" or the "Company") today announced its unaudited financial performance for the fourth quarter and full year ended December 31, 2025. As a btc-42">bitcoin mining enterprise relying on a globally operated layout and dedicated to building an integrated energy and AI computing power platform, CanGu is actively advancing its business transformation and infrastructure development.
• Financial Performance:
Total revenue for the full year 2025 was $688.1 million, with $179.5 million in the fourth quarter.
Bitcoin mining business revenue for the full year was $675.5 million, with $172.4 million in the fourth quarter.
Full-year adjusted EBITDA was $24.5 million, while the fourth quarter was -$156.3 million.
• Mining Operations and Costs:
A total of 6,594.6 bitcoins were mined throughout the year, averaging 18.07 bitcoins per day; of which 1,718.3 bitcoins were mined in the fourth quarter, averaging 18.68 bitcoins per day.
The average mining cost for the full year (excluding miner depreciation) was $79,707 per bitcoin, and for the fourth quarter, it was $84,552;
The all-in sustaining costs were $97,272 and $106,251 per bitcoin, respectively.
As of the end of December 2025, the company has cumulatively produced 7,528.4 bitcoins since entering the bitcoin mining business.
• Strategic Progress:
The company has completed the termination of the American Depositary Receipt (ADR) program and transitioned to a direct listing on the NYSE to enhance information transparency and align with its strategic direction, with a long-term goal of expanding its investor base.
CEO Paul Yu stated: "2025 marked the company's first full year as a bitcoin mining enterprise, characterized by rapid execution and structural reshaping. We completed a comprehensive adjustment of our asset system and established a globally distributed mining network. Additionally, the company introduced a new management team, further strengthening our capabilities and competitive advantage in the digital asset and energy infrastructure space. The completion of the NYSE direct listing and USD pricing also signifies our transformation into a global AI infrastructure company."
"As we enter 2026, the company will continue to optimize its balance sheet structure and enhance operational efficiency and cost resilience through adjustments to the miner portfolio. At the same time, we are advancing our strategic transformation into an AI infrastructure provider. Leveraging EcoHash, we will utilize our capabilities in scalable computing power and energy networks to provide cost-effective AI inference solutions. The relevant site transformations and product development are progressing simultaneously, and the company is well-positioned to sustain its execution in the new phase."
The company's Chief Financial Officer, Michael Zhang, stated: "By 2025, the company is expected to achieve significant revenue growth through its scaled mining operations. Despite recording a net loss of $452.8 million from ongoing operations, mainly due to one-time transformation costs and market-driven fair value adjustments, the company, from a financial perspective, will reduce its leverage, optimize its Bitcoin reserve strategy and liquidity management, introduce new capital to strengthen its financial position, and seize investment opportunities in high-potential areas such as AI infrastructure while navigating market volatility."
The total revenue for the fourth quarter was $1.795 billion. Of this, the Bitcoin mining business contributed $1.724 billion in revenue, generating 1,718.3 Bitcoins during the quarter. Revenue from the international automobile trading business was $4.8 million.
The total operating costs and expenses for the fourth quarter amounted to $4.56 billion, primarily attributed to expenses related to the Bitcoin mining business, as well as impairment of mining machines and fair value losses on Bitcoin collateral receivables.
This includes:
· Cost of Revenue (excluding depreciation): $1.553 billion
· Cost of Revenue (depreciation): $38.1 million
· Operating Expenses: $9.9 million (including related-party expenses of $1.1 million)
· Mining Machine Impairment Loss: $81.4 million
· Fair Value Loss on Bitcoin Collateral Receivables: $171.4 million
The operating loss for the fourth quarter was $276.6 million, a significant increase from a loss of $0.7 million in the same period of 2024, primarily due to the downward trend in Bitcoin prices.
The net loss from ongoing operations was $285 million, compared to a net profit of $2.4 million in the same period last year.
The adjusted EBITDA was -$156.3 million, compared to $2.4 million in the same period last year.
The total revenue for the full year was $6.881 billion. Of this, the revenue from the Bitcoin mining business was $6.755 billion, with a total output of 6,594.6 Bitcoins for the year. Revenue from the international automobile trading business was $9.8 million.
The total annual operating costs and expenses amount to $1.1 billion.
Specifically, they include:
· Revenue Cost (excluding depreciation): $543.3 million
· Revenue Cost (depreciation): $116.6 million
· Operating Expenses: $28.9 million (including related-party expenses of $1.1 million)
· Miner Impairment Loss: $338.3 million
· Bitcoin Collateral Receivable Fair Value Change Loss: $96.5 million
The full-year operating loss is $437.1 million. The continuing operations net loss is $452.8 million, while in 2024, there was a net profit of $4.8 million.
The 2025 non-GAAP adjusted net profit is $24.5 million (compared to $5.7 million in 2024). This measure does not include share-based compensation expenses; refer to "Use of Non-GAAP Financial Measures" for details.
As of December 31, 2025, the company's key assets and liabilities are as follows:
· Cash and Cash Equivalents: $41.2 million
· Bitcoin Collateral Receivable (Non-current, related party): $663.0 million
· Miner Net Value: $248.7 million
· Long-Term Debt (related party): $557.6 million
In February 2026, the company sold 4,451 bitcoins and repaid a portion of related-party long-term debt to reduce financial leverage and optimize the asset-liability structure.
As per the stock repurchase plan disclosed on March 13, 2025, as of December 31, 2025, the company had repurchased a total of 890,155 shares of Class A common stock for approximately $1.2 million.

The US AI Startup Is Loving China's Open Source Model

