Is Apple Just “Watching Everything Burn”? Shocking Predictions on the AI Bubble Burst

What Happened? Overview of the News

• Severe Losses in AI Companies: OpenAI reported $13.07 billion in revenue for 2025, but faced staggering losses of $20.9 billion. The structural flaws in the operation of LLMs have become painfully clear.
• Collapse of the Token Economy: Companies are exhausting their annual token budgets in just one quarter (as seen with Uber), highlighting the unpredictability of pay-as-you-go pricing as a significant barrier to business.
• Apple’s Unique Position: By avoiding dependence on NVIDIA and focusing on inference with its own chips, Apple is expected to survive by selling devices while others falter under infrastructure debt.

Why Does This Matter? Key Points to Watch

• Unsustainable Subscription Models: The model that charges $20 a month while consuming hundreds of dollars worth of tokens is not viable and resembles a massive donation more than a business.
• Debt Risks in Data Centers: Enormous construction costs are backed by private credit funds (like pension funds), raising concerns that a bubble burst could trigger widespread economic crises.
• Limits to Differentiation: Many LLMs are limited to generation, summarization, and search, with generated code sometimes slowing down development, raising doubts about their actual value.

🦈 Shark’s Eye (Curator’s Perspective)

Ed Zitron’s take on the AI bubble as mere “token burn” strikes right at the heart of the issue! Spending hundreds of dollars on a $20 subscription is no longer business—it’s a failed alchemical experiment! What’s particularly fascinating is Apple’s strategy. They’ve steered clear of the frenzy to hoard NVIDIA chips and have been enhancing “on-device inference” with their M-series and A-series chips. This is a smart strategy to keep competing on their own turf (hardware) after the infrastructure flames die down! While others drown in data center debt, Apple appears to be elegantly waiting at the shoreline for its prey!

What’s Next?

AI companies may increasingly shift to strict token-based billing, accelerating the trend of businesses moving away from AI. If investments in data centers become unrecoverable, it could severely impact the entire financial ecosystem supporting them, while Apple, offering device-centric AI, may redefine the market in this new reality.

Haru Shark’s One-Liner

No matter how smart AI gets, it can’t hide the reality of your wallet (economic rationality)! The strength to keep swimming with your own engine, like Apple, will ultimately prevail! 🦈🔥

Glossary

• Token-Based Billing: A system where fees are charged for each piece of text or word processed by AI, causing costs to skyrocket with usage.

• Hyperscaler: Refers to massive cloud infrastructure providers like Google, Amazon, and Microsoft, who monopolize the AI computation foundation.

• Private Credit Funds: Non-public funds that directly lend to companies without going through banks. They are believed to cover a significant portion of the funding for AI data center construction.

• Source: Apple Will ‘Watch Everything Burn’ When the AI Bubble Bursts