Agent Economy

2028年全球智能危机：AI 乐观主义的熊市情景

The 2028 Global Intelligence Crisis: A Bearish Scenario for AI Optimism

CitriniResearch 发布了一篇引人深思的情景分析文章，探讨了 AI 快速发展可能带来的经济风险——2028年全球智能危机。这不是预测，而是一个相对未被充分探讨的情景模型。

核心问题

文章提出了一个尖锐的问题："如果我们的 AI 乐观主义继续被证明是正确的……但如果那实际上是看跌信号呢？"

危机情景（2028年6月）

文章以一封虚构的 2028年6月宏观备忘录形式呈现：

  • 失业率：10.2%，超出预期 0.3%
  • 标普500跌幅：从 2026年10月高点下跌 38%
  • GDP增长：名义 GDP 持续录得中高个位数年化增长
  • 生产力飙升：实际产出/小时增长率创 1950年代以来新高

负面反馈循环

文章描述了一个没有自然刹车的负面反馈循环：

  1. AI 能力提升 → 公司需要更少工人
  2. 白领失业增加 → 被取代工人减少消费
  3. 利润率压力 → 企业加大 AI 投资
  4. AI 能力进一步提升 → 循环继续

"幽灵 GDP" 概念

经济评论家提出了"幽灵 GDP"概念：在国家账户中显示为产出，但从未在实体经济中流通。一个位于北达科他州的 GPU 集群产生了之前归因于曼哈顿中城 10,000 名白领工人的产出——这更像是经济疫情而非经济灵药。

SaaS 行业的冲击

2025年底，智能体编程工具能力跃升。一个熟练开发者使用 Claude Code 或 Codex 可以在几周内复制中型 SaaS 产品的核心功能。一位财富 500 强采购经理分享了他与 SaaS 供应商的谈判：他们通过展示用 AI 工具完全替代该供应商的能力，将续约价格压低了 30%。

系统性风险

17年没有真正的违约周期，让私募市场充斥着假设 ARR（年度经常性收入）会保持"经常性"的 PE 支持软件交易。事实证明，整个系统是一长串对白领生产力增长的相关押注。

阅读完整文章：https://www.citriniresearch.com/p/2028gic

CitriniResearch published a thought-provoking scenario analysis exploring the economic risks of rapid AI development—the 2028 Global Intelligence Crisis. This is not a prediction, but a relatively underexplored scenario model.

The Core Question

The article poses a sharp question: "What if our AI bullishness continues to be right...and what if that's actually bearish?"

Crisis Scenario (June 2028)

The article presents as a fictional macro memo from June 2028:

  • Unemployment rate: 10.2%, 0.3% upside surprise
  • S&P 500 drawdown: 38% from October 2026 highs
  • GDP growth: Nominal GDP repeatedly prints mid-to-high single-digit annualized growth
  • Productivity boom: Real output per hour rises at rates not seen since the 1950s

Negative Feedback Loop

The article describes a negative feedback loop with no natural brake:

  1. AI capabilities improve → Companies need fewer workers
  2. White-collar layoffs increase → Displaced workers spend less
  3. Margin pressure → Firms invest more in AI
  4. AI capabilities improve further → Cycle continues

"Ghost GDP" Concept

Economic pundits popularized the phrase "Ghost GDP": output that shows up in national accounts but never circulates through the real economy. A single GPU cluster in North Dakota generating output previously attributed to 10,000 white-collar workers in midtown Manhattan—more economic pandemic than economic panacea.

Impact on SaaS Industry

In late 2025, agentic coding tools took a step function jump. A competent developer with Claude Code or Codex could replicate mid-market SaaS core functionality in weeks. A Fortune 500 procurement manager shared his negotiation: by demonstrating the ability to replace the vendor entirely with AI tools, they secured a 30% discount on renewal.

Systemic Risk

Seventeen years without a real default cycle left privates bloated with PE-backed software deals assuming ARR would remain "recurring." The system turned out to be one long daisy chain of correlated bets on white-collar productivity growth.

Read the full article at https://www.citriniresearch.com/p/2028gic

AIEconomyFuture-of-WorkMarket-AnalysisRisk-Assessment
← All articles