NVIDIA and the Two Waves: From Crypto to AI — The Art of Riding a Bubble
🌊 Introduction: When Innovation Meets Speculation
In 2017 it was the crypto-mining gold rush; by 2023–2025 it became the AI data-center gold rush. At the center of both stands NVIDIA (NVDA). Is the company merely surfing these cycles—or subtly steering them?
💰 Phase 1 — Crypto Mining Boom (2017–2021)
During Ethereum’s Proof-of-Work era, NVIDIA gaming GPUs were diverted into mining, creating scarcity and outsized profits. In December 2024, the U.S. Supreme Court allowed a class-action suit to proceed that alleges NVIDIA misled investors about crypto-related sales exposure—keeping that question alive in the courts. (AP News)
🤖 Phase 2 — The AI Compute Rush (2022–Now)
As Ethereum moved to Proof-of-Stake in 2022, GPU mining demand collapsed—just as generative AI spiked. NVIDIA’s A100/H100 data-center GPUs became the default for training/inference at hyperscalers and model labs, while the CUDA / cuDNN / TensorRT stack deepened developer lock-in. Meanwhile, crypto miners began pivoting their power-rich sites into AI data centers, accelerating demand for AI-grade compute. (CoinDesk)
Recent deals underscore that shift: in October 2025, a consortium led by BlackRock—with NVIDIA, Microsoft, and xAI participating—announced a $40B acquisition of Aligned Data Centers to expand AI infrastructure capacity. Coverage from Reuters and Financial Times highlights the scale and intent; crypto/AI trade press note the valuation premium over listed bitcoin miners. (Reuters)
Miners → AI Operators (examples)
Industry reporting shows multiple miners (e.g., Bitdeer, Cipher) retooling for AI workloads, and analysts (Bernstein) quantifying miners’ advantaged access to power. (CoinDesk)
🧩 NVIDIA’s Playbook: Riding and Steering
| Lever | What it looks like | Result |
|---|---|---|
| Narrative timing | Transitioned messaging from gaming → crypto → AI right as each wave peaked/faded | Sustained growth story |
| Supply orchestration | Perceived scarcity of top GPUs (A/H-series) | Pricing power & ASPs |
| Ecosystem lock-in | CUDA + SDKs + systems partners | Switching costs rise |
| Story & signal | Headline growth + marquee partnerships & pre-buys | Confidence flywheel |
🧮 Two Waves, One Roadmap (Mermaid Gantt)
This version uses Gantt (widely supported) to avoid timeline-syntax issues.
gantt
title NVIDIA: Steering Two Waves — Crypto → AI
dateFormat YYYY
section Crypto Wave
Crypto mining boom (ETH PoW) :done, 2017, 2018
Crypto winter :crit, 2019, 2019
DeFi/NFT surge :active, 2020, 2021
section AI Wave
Ethereum Merge (PoS) :2022, 2022
GenAI breakout (ChatGPT launch) :2023, 2023
AI datacenter build-out :2024, 2024
Consolidation & scrutiny :2025, 2025
⚖️ Innovation vs. Exuberance
The AI build-out has real economics (productivity gains; hyperscaler capex), yet pockets of speculative demand persist (GPU hoarding; “AI wrapper” startups). Recent reporting compares the moment to the dot-com era—revolutionary tech with valuation heat. (Financial Times)
🧭 Bottom Line
NVIDIA isn’t committing illegal “market manipulation”; it’s executing market orchestration—aligning product cadence, ecosystem strategy, and partnerships to convert waves of enthusiasm into durable demand. Whether this resolves into permanent infrastructure—or another painful reset—depends on how quickly real AI workloads fill all that capacity. (CoinDesk)
📚 Updated, Working References
- Supreme Court / Lawsuit — AP News: Court allows class-action suit to proceed (Dec 11, 2024). (AP News)
- Background coverage — PBS NewsHour write-up (AP). (PBS)
- Hearing preview — AP: Court seemed likely to keep suit alive (Nov 13, 2024). (AP News)
- Crypto miners → AI data centers (analysis/reporting) — CoinDesk features & market notes (Sep–Oct 2025). (CoinDesk)
- Industry blog/analysis — DataCenters.com on miners pivoting to AI (Jul 21, 2025). (Data Centers)
- Podcast/industry press — Fierce Network: “Crypto companies mine for AI gold” (Oct 17, 2025). (Fierce Network)
- Aligned Data Centers deal — Reuters (Oct 15, 2025); Financial Times (Oct 15, 2025); The Block valuation context. (Reuters)
Get in Touch with us
Related Posts
- AI 反模式:AI 如何“毁掉”系统
- Anti‑Patterns Where AI Breaks Systems
- 为什么我们不仅仅开发软件——而是让系统真正运转起来
- Why We Don’t Just Build Software — We Make Systems Work
- 实用的 Wazuh 管理员 Prompt Pack
- Useful Wazuh Admin Prompt Packs
- 为什么政府中的遗留系统替换往往失败(以及真正可行的方法)
- Why Replacing Legacy Systems Fails in Government (And What Works Instead)
- Vertical AI Use Cases Every Local Government Actually Needs
- 多部门政府数字服务交付的设计(中国版)
- Designing Digital Service Delivery for Multi-Department Governments
- 数字政务服务在上线后失败的七个主要原因
- The Top 7 Reasons Digital Government Services Fail After Launch
- 面向市级与区级政府的数字化系统参考架构
- Reference Architecture for Provincial / Municipal Digital Systems
- 实用型 GovTech 架构:ERP、GIS、政务服务平台与数据中台
- A Practical GovTech Architecture: ERP, GIS, Citizen Portal, and Data Platform
- 为什么应急响应系统必须采用 Offline First 设计(来自 ATAK 的启示)
- Why Emergency Systems Must Work Offline First (Lessons from ATAK)
- 为什么地方政府的软件项目会失败 —— 如何在编写代码之前避免失败













