AI 生成内容的 SEO 政策与质量规则研究笔记
来源(全部为 Google 官方一手来源,引文逐字核对于 2026-08-02):
- 官方立场文章《Google Search’s guidance about AI-generated content》(2023-02-08,署名 Danny Sullivan & Chris Nelson,Google Search Quality team): https://developers.google.com/search/blog/2023/02/google-search-and-ai-content
- 垃圾内容政策(Scaled content abuse 一节): https://developers.google.com/search/docs/essentials/spam-policies
- 《Creating helpful, reliable, people-first content》(How/Why 段落): https://developers.google.com/search/docs/fundamentals/creating-helpful-content
- 2024-03 核心更新与新垃圾内容政策公告: https://developers.google.com/search/blog/2024/03/core-update-spam-policies
- 搜索质量评分者指南 PDF(2025-09-11 版,本目录已存档原件): https://static.googleusercontent.com/media/guidelines.raterhub.com/en//searchqualityevaluatorguidelines.pdf
AI 搜索功能(AI Overviews / AI Mode)的收录与优化规则另见 ai-search-features.md。
一、总原则:看质量,不看生产方式
2023-02 立场文章确立的核心原则,后续所有政策都从它推导:
“Our focus on the quality of content, rather than how content is produced, is a useful guide that has helped us deliver reliable, high quality results to users for years.”
文章用了一个历史类比:约十年前也出现过大量「人写的低质量量产内容」,Google 当时没有禁止所有人写内容,而是改进系统去奖励优质内容——“No one would have thought it reasonable for us to declare a ban on all human-generated content in response.”
同时划出红线(在 2023 博文、helpful-content 文档、2024-03 公告三处措辞一致):
“Using automation—including AI—to generate content with the primary purpose of manipulating ranking in search results is a violation of our spam policies.”
即:违规判定的枢纽是主要目的(primary purpose),不是是否使用 AI。官方同时指出并非所有自动化都是垃圾内容:体育比分、天气预报、文字转录长期由自动化生成。
二、官方 FAQ 全文整理(2023-02 立场文章,10 条)
| # | 问题 | 官方回答要点 |
|---|---|---|
| 1 | AI 内容违反指南吗? | 恰当使用不违反。「恰当」= 不以操纵排名为主要目的 |
| 2 | 为什么不禁止 AI 内容? | 出版业长期用自动化产出有用内容;AI 能以新方式辅助和生成有用内容 |
| 3 | 如何防止低质量 AI 内容占领搜索结果? | 低质量内容不是新问题;既有的实用性判断系统、原创报道提升系统持续改进 |
| 4 | 如何处理 AI 传播的错误信息? | 人写和 AI 内容都有此问题;健康、公民、金融等主题上系统更强调可靠性信号 |
| 5 | 如何识别 AI 制造的垃圾内容? | 「我们有多种系统,包括 SpamBrain,分析模式和信号来识别垃圾内容,不论其如何产出」 |
| 6 | AI 内容会排名靠前吗? | ”Using AI doesn’t give content any special gains. It’s just content.”——有用、原创、符合 E-E-A-T 就可能表现好,否则不会 |
| 7 | 我应该用 AI 生成内容吗? | 把 AI 当作产出有用原创内容的手段,可以考虑;当作低成本操纵排名的手段,不要 |
| 8 | 是否所有内容都要加作者署名? | 读者会想「这是谁写的?」时应加准确署名;Google News 出版方应有署名与作者信息 |
| 9 | 是否应添加 AI/自动化披露? | 「对读者可能会想『这是怎么做出来的?』的内容,披露有用;在合理预期的情况下考虑添加」 |
| 10 | 能把 AI 列为作者吗? | 「给 AI 署名作者大概不是向读者说明 AI 参与创作的最佳方式」——即不建议 |
三、Scaled Content Abuse(滥用规模化内容)政策全文
垃圾内容政策页的完整定义(2024-03 起的正式政策):
“Scaled content abuse is when many pages are generated for the primary purpose of manipulating search rankings and not helping users. This abusive practice is typically focused on creating large amounts of unoriginal content that provides little to no value to users, no matter how it’s created.”
官方示例(明确「包括但不限于」):
- 使用生成式 AI 工具或类似工具生成大量对用户无附加价值的页面
- 抓取 Feed、搜索结果或其他内容生成大量页面(包括同义词替换、机器翻译等自动化变换)
- 拼接不同网页的内容而无附加价值
- 创建多个网站以掩盖内容的规模化性质
- 创建大量对读者几乎无意义、只含搜索关键词的页面
对站长的要求(强制性措辞):“If you’re hosting such content on your site, exclude it from Search.”
与旧政策的关系(2024-03 公告 FAQ 原文)
这条政策取代了旧的「automatically-generated content」政策,官方解释了扩展原因:
“It’s been expanded to account for more sophisticated scaled content creation methods where it isn’t always clear whether low quality content was created purely through automation.”
以及:
“producing content at scale is abusive if done for the purpose of manipulating search rankings and that this applies whether automation or humans are involved.”
即政策从「针对自动生成」改为「针对规模化滥用」,人肉内容农场同样违规。处罚:“Sites that violate our spam policies may rank lower in results or not appear in results at all”;受手动操作(manual action)影响的站长会收到 Search Console 通知,可申请重新审核。
四、评分者指南(2025-09-11 版)中的 AI 内容评分规定
以下引文逐字取自官方 PDF(本目录存档件,已 MD5 核对与官方 URL 一致)。
§2.1 生成式 AI 的官方定义
“Generative AI is a type of machine learning (ML) model that can take what it has learned from the examples it has been provided to create new content, such as text, images, music, and code. … Generative AI can be a helpful tool for content creation, but like any tool, it can also be misused.”
§3.2 投入(Effort)的定义——质量判定的基准概念
“Effort: Consider the extent to which a human being actively worked to create satisfying content. … the automatic creation of thousands of pages by running existing freely available content through existing translation software without any oversight, manual curation, etc., would not be considered to have effort.”
§4.6.5 规模化内容滥用 → 评 Lowest
“Pages and websites made up of content created at scale with no original content or added value for users, should be rated Lowest, no matter how they are created. Even if you are unsure of the method of creation, e.g. whether or not the page is created using generative AI tools, you should still use the Lowest rating when you strongly suspect scaled content abuse after looking at several pages on the website.”
即:即使无法确认是否用了 AI,只要强烈怀疑规模化滥用即评 Lowest。
§4.6.6 关键平衡句:AI 工具本身不决定评级
“The Lowest rating applies if all or almost all of the MC on the page (including text, images, audio, videos, etc) is copied, paraphrased, embedded, auto or AI generated, or reposted from other sources with little to no effort, little to no originality, and little to no added value for visitors to the website.”
“Likewise, the use of Generative AI tools alone does not determine the level of effort or Page Quality rating. Generative AI tools may be used for high quality and low quality content creation. For example, a high level of effort may be involved in creating high quality original artwork using Generative AI tools.”
§4.6.7 识别 AI 改写内容的线索
转述/摘要式内容(paraphrased content)的特征之一:
“Have words or other indications of summarizing or paraphrasing generative AI tools, such as words like ‘As an AI language model’“
§4.7 官方给出的两个 Lowest 实例
- 一个 YMYL 医疗文章页,判定 Lowest(scaled content abuse),理由:文章开头是 “As a language model, I don’t have real-time data and my knowledge cutoff date is September 2021.”
- 一个养鸡问答页:虽然「内容创建方式未知」,但问答格式常被用于生成式 AI 低投入改写内容,且「文本看起来未经人类编辑撰写或编辑」。
§5.2 Low 档(未到滥用程度的低投入内容)
“MC is Low quality if it is created without adequate effort, originality, talent, or skill necessary to achieve the purpose of the page in a satisfying way.”
版本史注:生成式 AI 的定义与相关 Lowest 规定于 2025-01-23 版指南首次加入(此时间点来自第三方转述 Search Engine Land,官方 Change Log 的对应措辞是「Lowest/Low 章节与垃圾内容政策对齐」);现行措辞以 2025-09-11 官方 PDF 为准。第三方流传的短语 “created with little to no human effort” 在官方 PDF 中无逐字对应,官方措辞是 “little to no effort”,且明确 “means little to no effort of any type”。
五、披露(Disclosure)规则的准确边界
《Creating helpful content》“How” 一节的完整规定——当自动化被实质性地用于生成内容(substantially generate content)时,官方给出三个自问:
- 自动化/AI 的使用对访问者是否显而易见(通过披露或其他方式)?
- 是否提供了自动化/AI 如何被用于创建内容的背景?
- 是否解释了为什么自动化/AI 对产出内容有用?
判断标准:
“AI or automation disclosures are useful for content where someone might think ‘How was this created?’ Consider adding these when it would be reasonably expected.”
准确定性:披露的措辞是 “Consider adding”、“useful”,属于建议而非强制;官方从未说不披露即违规。但把 AI 列为作者被官方不建议(FAQ 第 10 条)。
同页的警示信号自查项:“Are you using extensive automation to produce content on many topics?”(是否在用大规模自动化跨大量主题产出内容)。
六、三层结构总结
| 层次 | 内容 |
|---|---|
| 允许 | 用 AI 产出内容本身不违规;AI 内容无加成也无扣分,与人写内容同一标准竞争;评分者明确「仅使用 AI 工具不决定评级」;高投入的 AI 原创(如原创艺术作品)被官方认可 |
| 建议 | 读者会问「这是怎么做出来的?」时加 AI 披露;说明如何用、为何用;加准确的人类署名;不给 AI 署名作者 |
| 惩罚 | 以操纵排名为主要目的的 AI/自动化内容违反垃圾内容政策;规模化内容滥用(无论 AI、人工或混合)→ 降位或完全移出结果 + 可能的手动操作;评分者对规模化滥用及「几乎全部主内容系复制/改写/AI 生成且无投入无原创无附加值」的页面一律评 Lowest |
一句话:Google 的政策变量从来不是「AI 与否」,而是「投入(effort)、原创性(originality)、附加价值(added value)、目的(purpose)」四项。 AI 只是把这四项做差的成本降到了极低,所以 2024-03 把政策从「针对自动生成」重写为「针对规模化滥用」。