2026
4 篇文章Debian 公開決議：LLM 貢獻的去留之爭
Debian 於 7 月 23 日起為「LLM 使用」啟動一般決議討論：八個提案從全面禁止到有條件開放，8 月 15 日開始投票，將為大型開源專案的 AI 貢獻治理立下先例。
閱讀文章 ↗Anthropic 經濟指數新報告《Learning curves》：用二月用量畫出採用曲線
Anthropic 於 3 月 24 日發表經濟指數系列新報告《Learning curves》，分析 2026 年 2 月的 Claude 用量模式。本文介紹這個系列的方法論價值與侷限，以及產品與招募團隊可以怎麼使用這份資料。
閱讀文章 ↗Anthropic 提出「實測暴露」：AI 能力與實際使用的三倍落差
Anthropic 於 2026 年 3 月 5 日發表勞動市場研究，提出「實測暴露」指標：結合 O*NET 任務資料與 Claude 實際用量。電腦與數學類任務理論上 94% 可被 AI 執行、實際僅 33% 被覆蓋；程式設計師以 75% 居首，22–25 歲求職率出現約 14% 的下滑訊號。
閱讀文章 ↗Anthropic 經濟指數第四份報告：56% 對話屬增強，41% 屬自動化
2026 年 1 月 15 日，Anthropic 發布第四份 Economic Index 報告，引入「經濟原型」分類：56% 的 Claude 對話屬於增強型、41% 屬於自動化型。本文解讀數字背後的勞動市場訊號與方法論限制。
閱讀文章 ↗
2026
5 ARTICLESDebian's General Resolution on LLM Contributions
Debian opened discussion on July 23 for a General Resolution on LLM usage: eight proposals spanning an outright ban to conditional acceptance, with voting from August 15 to 28.
READ POST ↗OpenRouter Classifiers: Making Every AI Request Account for Itself
OpenRouter's beta Classifiers tag every generation with structured metadata: a four-part config, six templates, async execution, zero added latency, and sampling-rate cost control.
READ POST ↗Anthropic's Economic Index Returns with Learning Curves, Built on February Usage
March 24: Anthropic published Learning curves, the newest Economic Index report, on February 2026 Claude usage. On the method, its limits, and its use for product and hiring teams.
READ POST ↗Anthropic Measures the Gap Between AI Ability and Real Use
Anthropic's new 'observed exposure' paper weighs real Claude usage against capability: programmers reach 75% task coverage, while 33% actual use trails 94% feasibility in computer and math work.
READ POST ↗Anthropic's Fourth Economic Index: 56% Augmentation, 41% Automation
Anthropic's fourth Economic Index report, published January 15, 2026, introduces economic primitives: 56% of Claude conversations classified as augmentation, 41% as automation. How to read the split.
READ POST ↗