{
 "generated": "2026-11-20",
 "sha": "",
 "node": {
  "date": "2026-07-15",
  "updated": "2026-07-15",
  "kind": "report",
  "category": "L5 · 全球地缘",
  "group": "governance",
  "page": "/deepdive/ai-model-ideology/",
  "summary": "Anthropic Diff Tool 首次量化 Llama/Qwen3-8B/DeepSeek 预训练权重中的意识形态特征，叠加 DeepSeek 为华为昇腾芯片延迟 V4 发布的硬件主权决策——「意识形态可测量化」与「算力主权物理化」两条新疆域同时打开",
  "version": 1,
  "tags": [
   "ai-geopolitics",
   "ideology",
   "deepseek",
   "huawei",
   "compute-sovereignty"
  ],
  "langs": [
   "zh",
   "en"
  ],
  "words": 3698,
  "n_sections": 4,
  "dd_id": "DD·0039",
  "related": [
   "open-source-encirclement",
   "ai-geopolitics-2025-2026",
   "capability-vs-compute",
   "liang-wenfeng-leak"
  ],
  "id": "dd:ai-model-ideology",
  "type": "Article",
  "article": "ai-model-ideology",
  "name": "AI 模型意识形态：可测量的地缘政治新疆域",
  "title": "AI 模型意识形态：可测量的地缘政治新疆域",
  "slug": "ai-model-ideology",
  "kind_zh": "深度报告",
  "entry": "agent",
  "n_ents": 10,
  "n_entities": 10
 },
 "page": "https://xiaopingfeng.com/graph/e/dd/ai-model-ideology/",
 "out": [
  {
   "type": "covers",
   "sentence": "AI 模型意识形态：可测量的地缘政治新疆域讲到DeepSeek",
   "direction": "out",
   "id": "app:deepseek",
   "name": "DeepSeek",
   "role": "mention",
   "status": "agent",
   "evidence": []
  },
  {
   "type": "covers",
   "sentence": "AI 模型意识形态：可测量的地缘政治新疆域讲到DeepSeek",
   "direction": "out",
   "id": "org:deepseek",
   "name": "DeepSeek（深度求索）",
   "role": "mention",
   "status": "agent",
   "evidence": []
  },
  {
   "type": "covers",
   "sentence": "AI 模型意识形态：可测量的地缘政治新疆域讲到Huawei",
   "direction": "out",
   "id": "org:huawei",
   "name": "Huawei（华为）",
   "role": "mention",
   "status": "agent",
   "evidence": []
  },
  {
   "type": "covers",
   "sentence": "AI 模型意识形态：可测量的地缘政治新疆域讲到Anthropic",
   "direction": "out",
   "id": "org:anthropic",
   "name": "Anthropic",
   "role": "mention",
   "status": "agent",
   "evidence": []
  },
  {
   "type": "covers",
   "sentence": "AI 模型意识形态：可测量的地缘政治新疆域讲到Meta",
   "direction": "out",
   "id": "org:meta",
   "name": "Meta",
   "role": "mention",
   "status": "agent",
   "evidence": []
  },
  {
   "type": "covers",
   "sentence": "AI 模型意识形态：可测量的地缘政治新疆域讲到Qwen",
   "direction": "out",
   "id": "app:qwen",
   "name": "Qwen",
   "role": "mention",
   "status": "agent",
   "evidence": []
  },
  {
   "type": "covers",
   "sentence": "AI 模型意识形态：可测量的地缘政治新疆域讲到Alibaba",
   "direction": "out",
   "id": "org:alibaba",
   "name": "Alibaba（阿里巴巴）",
   "role": "mention",
   "status": "agent",
   "evidence": []
  },
  {
   "type": "covers",
   "sentence": "AI 模型意识形态：可测量的地缘政治新疆域讲到ByteDance",
   "direction": "out",
   "id": "org:bytedance",
   "name": "ByteDance（字节跳动）",
   "role": "mention",
   "status": "agent",
   "evidence": []
  },
  {
   "type": "covers",
   "sentence": "AI 模型意识形态：可测量的地缘政治新疆域讲到NVIDIA",
   "direction": "out",
   "id": "org:nvidia",
   "name": "NVIDIA",
   "role": "mention",
   "status": "agent",
   "evidence": []
  },
  {
   "type": "covers",
   "sentence": "AI 模型意识形态：可测量的地缘政治新疆域讲到Tencent",
   "direction": "out",
   "id": "org:tencent",
   "name": "Tencent（腾讯）",
   "role": "mention",
   "status": "agent",
   "evidence": []
  }
 ],
 "in": [
  {
   "type": "part_of",
   "sentence": "意识形态的算法指纹：Anthropic Diff Tool 发现了什么属于AI 模型意识形态：可测量的地缘政治新疆域",
   "direction": "in",
   "id": "ddsec:ai-model-ideology-anthropic-diff-tool",
   "name": "意识形态的算法指纹：Anthropic Diff Tool 发现了什么",
   "status": "derived",
   "evidence": []
  },
  {
   "type": "part_of",
   "sentence": "硬件层的主权宣言：DeepSeek 为华为芯片延迟旗舰属于AI 模型意识形态：可测量的地缘政治新疆域",
   "direction": "in",
   "id": "ddsec:ai-model-ideology-deepseek",
   "name": "硬件层的主权宣言：DeepSeek 为华为芯片延迟旗舰",
   "status": "derived",
   "evidence": []
  },
  {
   "type": "part_of",
   "sentence": "对企业和决策者意味着什么：从「选模型」变成「选阵营」属于AI 模型意识形态：可测量的地缘政治新疆域",
   "direction": "in",
   "id": "ddsec:ai-model-ideology-s",
   "name": "对企业和决策者意味着什么：从「选模型」变成「选阵营」",
   "status": "derived",
   "evidence": []
  },
  {
   "type": "part_of",
   "sentence": "开源生态的信任重构：透明的代价与价值属于AI 模型意识形态：可测量的地缘政治新疆域",
   "direction": "in",
   "id": "ddsec:ai-model-ideology-s-2",
   "name": "开源生态的信任重构：透明的代价与价值",
   "status": "derived",
   "evidence": []
  }
 ]
}