Anthropic's IPO Looks Like a Speculative Disaster in the Making: Why I'm Staying Away

By healthranger // 2026-09-30

Mastodon

Parler

Gab

Copy

The Most Anticipated IPO in Decades

Anthropic is filing for an IPO at a reported valuation near $2 trillion while disclosing some frightening financials ^[1]. A leaked prospectus reportedly shows a $42 billion net loss, $518 billion in unfunded spending commitments, and only $20 billion in cash ^[1]. Don't take this as investment advice, but my position is simple: I would not put a dollar into this IPO, because I believe it is a wealth-destruction event dressed up as a tech triumph. The market is being asked to fund billions in cash burn with no clear path to lasting profitability, and I wouldn't want to be a bag holder for insiders looking to exit. The most revealing aspect of a $2 trillion Anthropic IPO is not the staggering valuation, but the suspension of disbelief required to justify it ^[2]. To buy into that number, investors must assume that artificial intelligence is becoming economically indispensable at breakneck speed, that a formidable moat will protect a handful of frontier-model developers, and that the exorbitant costs of computing infrastructure will not eventually crush their margins ^[2]. I do not accept those assumptions, and I think anyone who does is setting themselves up for catastrophic losses. The IPO boom is not a sign of market strength. It is an unprecedented wave of equity issuance that could overwhelm available investor capital ^[3]. Companies like SpaceX, Anthropic, and OpenAI are looking to IPO hundreds of billions in common stock, not counting the even more hundreds of billions in lock-up expirations that will hit the market soon ^[4]. I own no tech stocks, and I'm not shorting anything, but I have said before that the stock market has become a complete fraud, propped up by endless government intervention and central bank money printing ^[5].

A Fragile Customer Base and No Moat

Anthropic's business rests on a customer base so thin that it terrifies me. A full 24% of its revenue comes from only two customers using its AI inference services ^[1]. That is not a moat. That is a cliff. Those two customers could switch to a different model, take inference in-house, or move to a cheaper open-source alternative at any time, and when they do, nearly a quarter of Anthropic's top line evaporates overnight. Anthropic's premium pricing depends on being irreplaceable, but in my view its customers have too many other options. OpenAI is already considering drastic price cuts in pursuit of Anthropic's customers, and the AI price wars have clearly begun ^[6]. When your competitor can undercut you on price and your customers can replicate your product with open-source tools, you do not have pricing power. You have a temporary lead that's melting away. The company's own CEO, Dario Amodei, has issued dire warnings that AI-induced unemployment could spike, and he has called for a pause in AI development until hand-picked arbiters are installed inside the frontier labs ^[7] ^[8]. I find it telling that the man running the company is publicly warning about the dangers of his own product while simultaneously asking the public to hand him $2 trillion. That does not inspire confidence. There is also the question of government dependency. The Department of War has pushed to compel supply chain companies to reject Anthropic, labeling the company's AI as a supply chain risk. And Anthropic's refusal to integrate Claude into Pentagon autonomous systems has reportedly forced partners like Boeing to reconsider their relationships ^[9] ^[10]. When your largest revenue sources are a handful of hyperscalers and government contractors who can walk away at any moment, you are not building a durable business. You are renting one.

The Cash Burn Has No Obvious Bottom

Anthropic is burning billions of dollars a year, and I still do not see where a lasting revenue model comes from. For leading AI companies, the biggest expense is not talent. It is compute ^[11]. In every case, compute accounts for the majority of total spending, underscoring how capital-intensive it has become to build and serve frontier AI models ^[11]. Anthropic is committed to spending hundreds of billions it does not have, and the money has to come from somewhere. The company sells inference through subscriptions and per-token API use, but open-source models cost a fraction of that. Inference is commoditizing fast, so I believe the billions needed to sustain Anthropic's operations will not materialize from a defensible business model. The AI boom is not a technology cycle. It is a credit-driven real-estate-like cycle whose financing architecture depends on the second derivative ^[12]. Levels and growth rates are the only numbers anyone watches, and the second derivative is where the whole thing collapses. I have been warning about this for a while. I have previously reported that the AI stock melt-up sparked fears of a bubble burst, with some AI companies trading at 90x forward earnings while OpenAI and other AI labs projected billions in losses through 2028 ^[13] ^[14]. The pattern is always the same: insiders cash out, retail investors get left holding the bag, and the whole thing is dressed up as innovation. Companies like SpaceX, Anthropic, and OpenAI are being valued in the trillions, yet they mostly generate negligible profits and have no realistic path to earnings that justify those numbers ^[15]. Launching IPOs for these companies is not innovation. It is a transfer of wealth from the gullible public to insiders who have been waiting years for their exit ^[15]. SpaceX, for context, slid despite a big revenue beat on its first earnings as a public company, and the August 6 lockup expiration opened the door to the first meaningful wave of insider selling just days after results ^[16]. That is the future that awaits Anthropic's public shareholders. The insiders get liquidity. You get dilution and downside.

Open Source Is Already Good Enough

I used to do most of my coding with Claude Code, Opus, and Sonnet, but today I use Qwen Flash Next locally for terminal work, configuration, and system monitoring. The shift was not ideological. It was practical. The open-source models got good enough, and they cost a fraction of what Anthropic charges. I have said before that China has demonstrated its AI models are superior and extraordinary, with models like Qwen 3.5 using a mixture-of-experts architecture that activates only 3 billion parameters out of 35 billion to respond efficiently ^[17]. DeepSeek, Kimi K3, and GLM 5.3 are close enough to Anthropic's models for most real-world coding tasks, and they are vastly cheaper. I currently use DeepSeek V4-1 Flash daily, and I employ Open Code as the front-end ^[18]. Meanwhile, Anthropic's own models are hobbled by censorship and refusals that make them less useful for real work. The censorship regime within AI development is more pronounced in the United States than in China, as evidenced by AI platforms such as Anthropic refusing to address critical questions about Big Pharma, while DeepSeek has provided comprehensive explanations and resources ^[19]. Open source also protects company data, avoids the censorship and refusals common in Anthropic's models, and lets businesses project costs instead of guessing at future API price hikes. For around $10,000 to $12,000, you can set up a robust yet consumer-grade system on your desk and expose it over your network for use by the entire company ^[20]. That setup allows for more than just prompting. It facilitates analysis in areas like logistics, and it turns one person into five ^[20]. Why would any rational business pay Anthropic's premium when they can own the hardware and run nearly-as-good models locally? I think that China will win the race for AI supremacy as US efforts collapse under woke, irrational demands for AI censorship ^[21]. Nothing has changed my mind. If anything, the gap has widened. The open-source community is making significant strides in keeping pace with proprietary corporate AI, and China is developing some of the most impressive models, such as DeepSeek, which are far more cost-effective ^[22]. Anthropic's moat is not technology. It is marketing.

Conclusion: The AI Bubble Will Pop, Don't Get Wrecked

I believe Anthropic will eventually collapse or be acquired, and OpenAI may fall around the same time. Both companies filed confidentially for IPO within a week of each other, and both are racing to extract as much capital as possible before the music stops ^[23]. When the correction comes, it will not be gentle. Companies that burn billions with no path to profitability do not get slow landings. They get repriced overnight. Because the stock market is now heavily AI-adjacent, that collapse could drag down the broader market. The equity supply surge is meeting a market priced for perfection, and history is not kind to that setup ^[24]. I have been watching the latest wave of tech IPOs with a sense of dread, not excitement, and I cannot wait for the AI bubble to crash because the clearing of malinvestment will open the door to a saner economy built on real value ^[15] ^[25]. Watch the bubble closely, and prepare for the fallout. Stack gold and silver. Decentralize your knowledge and your money. Use tools like BrightAnswers.ai for uncensored research, BrightLearn.ai to educate yourself for free, and Brighteon.social to stay connected without Big Tech censorship. The people who survive the coming reset will be the ones who saw it coming and positioned themselves accordingly.

References

Leaked Anthropic IPO Prospectus Shows $42BN Net Loss, $518BN In Unfunded Spending Commitments, And $20BN In Cash. - ZeroHedge. September 29, 2026.
Anthropic's $2 Trillion Bet Is That Nothing Goes Wrong. - ZeroHedge. August 13, 2026.
This Chart Should Stop You Cold In Your Tracks'. - ZeroHedge. June 15, 2026.
Ringing The Bell: Meta Plunges On Report It May Sell 'Tens Of Billions' In New Stock. - ZeroHedge. June 5, 2026.
The Great Stock Market Illusion: Why I'm Warning You to Stay Away from These Overpriced IPOs. - NaturalNews.com. June 5, 2026.
AI Price Wars Begin: OpenAI Considers 'Drastic Price Cuts' In Pursuit Of Anthropic Customers. - ZeroHedge. June 11, 2026.
AI Apocalypse: CEOs Warn of Job Destruction, Rogue AI Threatens Humanity. - NaturalNews.com. Patrick Lewis. March 15, 2026.
Zuck Torches Dario's AI Nanny State, Wants 'Trust Us Bro' Instead. - ZeroHedge. September 16, 2026.
Bright Videos Network - Anthropic Trump Dorsey Zuck transcript - BrightVideos.com. February 27, 2026.
Bright Videos News - Interview with Chris Martenson - Mike Adams - BrightVideos.com. March 5, 2026.
Compute Costs More Than Talent In AI. - ZeroHedge. April 27, 2026.
Forget CDOs, Meet CCOs: This Isn't A Tech Cycle... It's 2008 With Silicon. - ZeroHedge. August 14, 2026.
AI Stock Melt-Up Sparks Fears of 2025 Bubble Burst. - NaturalNews.com. Willow Tohi. October 17, 2025.
Nvidia's Explosive Earnings Defy AI Bubble Worries, Fueling Market Rally. - NaturalNews.com. November 21, 2025.
IPOs Are a Bag-Holder Scam: Why I'm Buying Gold Instead of Overpriced Tech Stocks. - NaturalNews.com. June 9, 2026.
SpaceX Slides Despite Big Revenue Beat, Tame CapEx. - ZeroHedge. August 4, 2026.
Bright Videos Network - China Anthropic transcript - BrightVideos.com. February 26, 2026.
Bright Videos News - Interview with Aaron Day transcript - Mike Adams - BrightVideos.com. May 22, 2026.
Brighteon Broadcast News - WORSHIP Of Vaccines - Mike Adams - Brighteon.com. January 31, 2025.
AIs Golden Promises and Dark Risks With Mike Adams - PeakProsperity.com. Chris Martenson. March 6, 2026.
Why China Will Win the Race for AI Supremacy as US Efforts Collapse Under Woke, Irrational Demands for AI Censorship. - NaturalNews.com. Mike Adams. January 6, 2025.
Brighteon Broadcast News - AI DOMINANCE - Mike Adams - Brighteon.com. January 22, 2025.
Following Anthropic, OpenAI files confidentially for IPO. - TechCrunch. June 8, 2026.
Equity Supply Surge: What Historically Comes Next. - ZeroHedge. Lance Roberts. June 8, 2026.
Why I Can't Wait for the AI Bubble to Crash. - NaturalNews.com. June 8, 2026.

Explainer Infographic