AI 垃圾研究證據輯

Presenting different content to users and search engines with the intent to manipulate search rankings and mislead users.

對使用者與搜尋引擎呈現不同內容，意圖操縱搜尋排名並誤導使用者。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

Sites or pages are created to rank for specific, similar search queries. They lead users to intermediate pages that are not as useful as the final destination.

為特定、相似的搜尋字詞建立網站或頁面，把使用者導向不如最終目的地有用的中介頁。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

An expired domain name is purchased and repurposed primarily to manipulate search rankings by hosting content that provides little to no value to users.

購入過期網域並改作他用，主要目的是掛上對使用者幾無價值的內容來操縱排名。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

Any content placed on a site without permission, due to vulnerabilities in a site's security.

因網站安全漏洞而在未經許可下被放上網站的任何內容。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

The practice of placing content on a page in a way solely to manipulate search engines and not to be easily viewable by human visitors.

把內容放在頁面上但只給搜尋引擎看、讓真人訪客不易看見的做法。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

Filling a web page with keywords or numbers in an attempt to manipulate rankings in Google Search results. Often these keywords appear in a list or group, unnaturally, or out of context.

在頁面塞滿關鍵字或數字以操縱排名，常以清單、成群、不自然或脫離語境的方式出現。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

The practice of creating links to or from a site primarily for the purpose of manipulating search rankings.

主要為操縱搜尋排名而建立指向或來自某網站的連結。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

The practice of sending automated queries to Google. This includes scraping results for rank-checking purposes or other types of automated access to Google Search conducted without express permission.

向 Google 發送自動化查詢，包括為查排名而抓取結果，或其他未經明確許可的自動化存取。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

Create a mismatch between user expectations and the actual outcome, leading to a negative and deceptive user experience, or compromised user security or privacy.

造成使用者預期與實際結果不符，帶來負面、欺騙性的體驗，或危及使用者安全與隱私。

Google — Spam policies for Google web search（2026-05-15 版；2026-04-13 官方公告將 back button hijacking 明列為本類目違規）｜developers.google.com/search/blog/2026/04/back-button-hijacking｜抓取 2026-08-11

Intentionally creating sites that trick users into thinking they would be able to access some content or services but in reality can't.

刻意建立讓使用者以為能取得某內容或服務、實際上卻拿不到的網站。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

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.

為操縱排名而非幫助使用者而大量產頁；典型是大量無原創性、對使用者幾無價值的內容——不論用什麼方式產生。官方例示包括「用生成式 AI 工具大量產頁而不為使用者加值」。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies#scaled-content｜抓取 2026-08-11

The practice of taking content from other sites, often through automated means, and hosting it with the purpose of manipulating search rankings.

常以自動化手段搬取他站內容並掛在自己網站上，以操縱搜尋排名。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

A tactic where third-party content is published on a host site mainly because of that host's already-established ranking signals, which it has earned primarily from its first-party content. The goal of this tactic is for the content to rank better than it could otherwise on its own.

把第三方內容掛在信譽已建立的網站上，借用該站的排名訊號，使內容排得比它自己能排的更前面。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

The practice of doing this maliciously in order to either show users and search engines different content or show users unexpected content that does not fulfill their original needs.

惡意轉址：讓使用者與搜尋引擎看到不同內容，或讓使用者看到無法滿足其原始需求的意外內容。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

The practice of publishing content with product affiliate links where the product descriptions and reviews are copied directly from the original merchant without any original content or added value.

掛聯盟連結但商品描述與評論直接照抄原商家，毫無原創內容或附加價值。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

Spammy content added to a site by users through a channel intended for user content. Often site owners are unaware of the spammy content.

使用者透過 UGC 管道加到網站上的垃圾內容，站主往往不知情。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

When Google receives significant volume of valid copyright removal requests involving a site, they demote other content from that site. Similar demotion applied to defamation, counterfeit goods, and court-ordered removal complaints. CSAM content always removed; sites with significant CSAM proportions have all content demoted.

網站累積大量有效版權移除請求時，該站其他內容會被降級；誹謗、仿冒品、法院命令移除同理；CSAM 一律移除且占比高的網站全站降級。

Google — Spam policies for Google web search（2026-05-15 版，降級類目）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

Demoting sites with significant volume of exploitative removal practices. Similar demotion for sites receiving significant removals for doxxing content, explicit non-consensual intimate imagery, or explicit non-consensual fake content (deepfakes).

對累積大量剝削式移除請求的網站降級；肉搜、未經同意的私密影像、未經同意的深偽內容同理。

Google — Spam policies for Google web search（2026-05-15 版，降級類目）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

If a site continues to engage in actions intended to bypass our spam policies or content policies for Google Search, we may take appropriate action which may include restricting or removing eligibility for some of our search features (for example, Top Stories, Discover) and taking broader action in Google Search (for example, removing more sections of a site from Search results).

若網站持續規避垃圾或內容政策，Google 可限制或取消其搜尋功能資格（如 Top Stories、Discover），並在搜尋中採取更大範圍的行動。

Google — Spam policies for Google web search（2026-05-15 版，執法升級條款）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

Come in many forms, including but not limited to impersonating an official business or service through imposter sites, intentionally displaying false information about a business or service, or otherwise attracting users to a site on false pretenses.

形式多樣：假冒官方企業或服務的仿冒站、刻意顯示不實的商家資訊、或以其他不實藉口吸引使用者。

Google — Spam policies for Google web search（2026-05-15 版）｜developers.google.com/search/docs/essentials/spam-policies｜抓取 2026-08-11

Google's ranking systems aim to reward original, high-quality content that demonstrates qualities of what we call E-E-A-T: expertise, experience, authoritativeness, and trustworthiness.

Google 排名系統旨在獎勵展現 E-E-A-T（專業、經驗、權威、可信）的原創高品質內容。

Google Search Central Blog — Google Search's guidance about AI-generated content（2023-02）｜developers.google.com/search/blog/2023/02/google-search-and-ai-content｜抓取 2026-08-10

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.

Google 的判準是內容品質、而非內容怎麼產生的。

Google Search Central Blog — Google Search's guidance about AI-generated content（2023-02）｜developers.google.com/search/blog/2023/02/google-search-and-ai-content｜抓取 2026-08-10

Using automation—including AI—to generate content with the primary purpose of manipulating ranking in search results is a violation of our spam policies.

用自動化（含 AI）產內容、主要目的是操縱搜尋排名，即違反垃圾政策。

Google Search Central Blog — Google Search's guidance about AI-generated content（2023-02）｜developers.google.com/search/blog/2023/02/google-search-and-ai-content｜抓取 2026-08-10

not all use of automation, including AI generation, is spam. Automation has long been used to generate helpful content, such as sports scores, weather forecasts, and transcripts.

並非一切自動化（含 AI 生成）都是垃圾；自動化早已用來產生有用內容，如賽事比分、天氣預報、逐字稿。

Google Search Central Blog — Google Search's guidance about AI-generated content（2023-02）｜developers.google.com/search/blog/2023/02/google-search-and-ai-content｜抓取 2026-08-10

Appropriate use of AI or automation is not against our guidelines. This means that it is not used to generate content primarily to manipulate search rankings, which is against our spam policies.

適當使用 AI 或自動化不違反指南；「適當」的定義是不以操縱排名為主要目的。

Google Search Central Blog — 同文 FAQ（2023-02）｜developers.google.com/search/blog/2023/02/google-search-and-ai-content｜抓取 2026-08-10

Poor quality content isn't a new challenge for Google Search to deal with. We've been tackling poor quality content created both by humans and automation for years.

低品質內容不是新挑戰；Google 多年來一直在處理人寫的與自動產的低品質內容。

Google Search Central Blog — 同文 FAQ（2023-02）｜developers.google.com/search/blog/2023/02/google-search-and-ai-content｜抓取 2026-08-10

We have a variety of systems, including SpamBrain, that analyze patterns and signals to help us identify spam content, however it is produced.

Google 有多套系統（含 SpamBrain）分析模式與訊號來辨識垃圾內容，不論其產製方式。

Google Search Central Blog — 同文 FAQ（2023-02）｜developers.google.com/search/blog/2023/02/google-search-and-ai-content｜抓取 2026-08-10

Using AI doesn't give content any special gains. It's just content. If it is useful, helpful, original, and satisfies aspects of E-E-A-T, it might do well in Search. If it doesn't, it might not.

用 AI 不會給內容任何加分；有用、原創、符合 E-E-A-T 就可能排得好，反之則否。

Google Search Central Blog — 同文 FAQ（2023-02）｜developers.google.com/search/blog/2023/02/google-search-and-ai-content｜抓取 2026-08-10

If you see AI as an essential way to help you produce content that is helpful and original, it might be useful to consider. If you see AI as an inexpensive, easy way to game search engine rankings, then no.

若把 AI 當成產出有用原創內容的必要幫手，可以考慮；若當成便宜取巧的排名工具，就不要。

Google Search Central Blog — 同文 FAQ（2023-02）｜developers.google.com/search/blog/2023/02/google-search-and-ai-content｜抓取 2026-08-10

Giving AI an author byline is probably not the best way to follow our recommendation to make clear to readers when AI is part of the content creation process.

把 AI 列為署名作者，大概不是向讀者交代 AI 參與產製的最佳方式。

Google Search Central Blog — 同文 FAQ（2023-02）｜developers.google.com/search/blog/2023/02/google-search-and-ai-content｜抓取 2026-08-10

Generative AI can be particularly useful when researching a topic, and to add structure to original content. However, using generative AI tools or other similar tools to generate many pages without adding value for users may violate Google's spam policy on scaled content abuse.

生成式 AI 在研究主題、為原創內容加結構時特別有用；但用它大量產頁而不為使用者加值，可能違反規模化內容濫用政策。

Google — Guidance on using generative AI content（最後更新 2025-12-10 UTC）｜developers.google.com/search/docs/fundamentals/using-gen-ai-content｜抓取 2026-08-10

If you're using generative AI content on your website, make sure your work meets the standards of the Search Essentials and our spam policies.

網站使用生成式 AI 內容時，必須符合 Search Essentials 與垃圾政策的標準。

Google — Guidance on using generative AI content（最後更新 2025-12-10 UTC）｜developers.google.com/search/docs/fundamentals/using-gen-ai-content｜抓取 2026-08-10

When creating content for the web, focus on accuracy, quality, and relevance, especially when automatically generating the content. This includes metadata like <title> elements, meta description elements, structured data, and alternate texts for images, which can appear in Search results.

產製網頁內容——尤其自動生成時——要聚焦準確性、品質與相關性，連 title、meta description、結構化資料、圖片替代文字等中繼資料都算。

Google — Guidance on using generative AI content（最後更新 2025-12-10 UTC）｜developers.google.com/search/docs/fundamentals/using-gen-ai-content｜抓取 2026-08-10

Sharing information about how a piece of content was created can help give your readers more context. If you're automatically generating content, consider adding information on how your content was created in a way that makes sense for your audience, such as by providing more background information on how automation was used and adding image metadata.

交代內容怎麼產生的能給讀者脈絡；自動生成時，宜以受眾看得懂的方式補充自動化使用背景並加上圖片中繼資料。

Google — Guidance on using generative AI content（最後更新 2025-12-10 UTC）｜developers.google.com/search/docs/fundamentals/using-gen-ai-content｜抓取 2026-08-10

For ecommerce sites, Google Merchant Center has policies for AI-generated content. In particular, AI-generated images must contain metadata using the IPTC DigitalSourceType TrainedAlgorithmicMedia metadata. AI-generated product data such as title and description attributes must be specified separately and labeled as AI-generated.

電商網站：AI 生成圖片必須帶 IPTC DigitalSourceType TrainedAlgorithmicMedia 中繼資料；AI 生成的商品標題與描述必須另行指定並標示為 AI 生成。

Google — Guidance on using generative AI content（最後更新 2025-12-10 UTC）｜developers.google.com/search/docs/fundamentals/using-gen-ai-content｜抓取 2026-08-10

You might find value in looking at the Search Quality Raters guidelines on how to evaluate both scaled content abuse (section 4.6.5) and main content created with little to no effort, little to no originality, and little to no added value (section 4.6.6). These guidelines are not a guide to ranking first in Google; they're used by our search raters to help evaluate the performance of our various search ranking systems, and their ratings don't directly influence ranking.

QRG 的 4.6.5（規模化內容濫用）與 4.6.6（低努力低原創低價值主內容）值得參考；但 QRG 不是排名攻略，評分員的評分不直接影響排名。

Google — Guidance on using generative AI content（最後更新 2025-12-10 UTC）｜developers.google.com/search/docs/fundamentals/using-gen-ai-content｜抓取 2026-08-10

People-first content means content that's created primarily for people, and not to manipulate search engine rankings.

People-first 內容＝主要為人而做、不是為操縱搜尋排名而做的內容。

Google — Creating helpful, reliable, people-first content（最後更新 2025-12-10 UTC）｜developers.google.com/search/docs/fundamentals/creating-helpful-content｜抓取 2026-08-10

If you use automation, including AI-generation, to produce content for the primary purpose of manipulating search rankings, that's a violation of our spam policies.

用自動化（含 AI 生成）產內容、主要目的是操縱搜尋排名，即違反垃圾政策。

Google — Creating helpful, reliable, people-first content（最後更新 2025-12-10 UTC）｜developers.google.com/search/docs/fundamentals/creating-helpful-content｜抓取 2026-08-10

Is the use of automation, including AI-generation, self-evident to visitors through disclosures or in other ways?

自我檢核題：自動化（含 AI 生成）的使用，是否透過揭露或其他方式讓訪客一目了然？

Google — Creating helpful, reliable, people-first content（最後更新 2025-12-10 UTC）｜developers.google.com/search/docs/fundamentals/creating-helpful-content｜抓取 2026-08-10

Of these aspects, trust is most important. The others contribute to trust, but content doesn't necessarily have to demonstrate all of them.

E-E-A-T 各面向中，信任（trust）最重要；其他面向是為信任服務，內容不必四項全備。

Google — Creating helpful, reliable, people-first content（最後更新 2025-12-10 UTC）｜developers.google.com/search/docs/fundamentals/creating-helpful-content｜抓取 2026-08-10

While E-E-A-T itself isn't a specific ranking factor, using a mix of factors that can identify content with good E-E-A-T is useful.

E-E-A-T 本身不是特定排名因子，但系統會用一組能辨識好 E-E-A-T 內容的因子。

Google — Creating helpful, reliable, people-first content（最後更新 2025-12-10 UTC）｜developers.google.com/search/docs/fundamentals/creating-helpful-content｜抓取 2026-08-10

LLMS.txt files and other "special" markup: You don't need to create new machine readable files, AI text files, markup, or Markdown to appear in Google Search (including its generative AI capabilities), as Google Search itself doesn't use them.

不需要建立 llms.txt、AI 專用檔案或標記才能出現在 Google 搜尋（含生成式 AI 功能）——Google 搜尋本身不使用它們。

Google — AI optimization guide（2026-05-15 加入；頁面最後更新 2026-07-10）｜developers.google.com/search/docs/fundamentals/ai-optimization-guide｜抓取 2026-08-10

"Chunking" content: There's no requirement to break your content into tiny pieces for AI to better understand it.

沒有任何要求叫你把內容切成小塊好讓 AI 理解。

Google — AI optimization guide（2026-05-15 加入；頁面最後更新 2026-07-10）｜developers.google.com/search/docs/fundamentals/ai-optimization-guide｜抓取 2026-08-10

Rewriting content just for AI systems: You don't need to write in a specific way just for generative AI search.

不需要為生成式 AI 搜尋用特定方式寫作。

Google — AI optimization guide（2026-05-15 加入；頁面最後更新 2026-07-10）｜developers.google.com/search/docs/fundamentals/ai-optimization-guide｜抓取 2026-08-10

The best practices for SEO continue to be relevant because our generative AI features on Google Search are rooted in our core Search ranking and quality systems.

SEO 最佳實務仍然適用，因為 Google 搜尋的生成式 AI 功能植根於核心排名與品質系統。

Google — AI optimization guide（2026-05-15 加入；頁面最後更新 2026-07-10）｜developers.google.com/search/docs/fundamentals/ai-optimization-guide｜抓取 2026-08-10

Don't just recycle what others on the internet have already said, or could easily be produced by a generative AI model.

別只是回收網路上別人已說過、或生成式 AI 模型輕易就能產出的內容。

Google — AI optimization guide（2026-05-15 加入；頁面最後更新 2026-07-10）｜developers.google.com/search/docs/fundamentals/ai-optimization-guide｜抓取 2026-08-11

ignore tactics like ... pursuing inauthentic mentions

指南摘要原文：忽略諸如「追逐不真實提及（pursuing inauthentic mentions）」之類的手法。注意：這是官方勸避免的 AEO/GEO 手法描述，不是垃圾政策類目，也不是獨立違規名稱（詳見第 10 節勘誤 10.3）。

Google — AI optimization guide 摘要（2026-05-15 加入；頁面最後更新 2026-07-10）｜developers.google.com/search/docs/fundamentals/ai-optimization-guide｜抓取 2026-08-11

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. Learn more here. Different tools leverage these models to create generative AI content. Generative AI can be a helpful tool for content creation, but like any tool, it can also be misused.

生成式 AI 是能從範例中學習並創造新內容（文字、圖片、音樂、程式碼）的機器學習模型；它可以是有用的創作工具，但如同任何工具，也可能被濫用。

Google — Search Quality Rater Guidelines §2.1 Important Definitions（2025-09-11 版）｜官方 PDF｜抓取 2026-08-10

Scaled content abuse is a spam practice described in the Google Search Web Spam Policies. Scaled content abuse occurs when many pages are generated for the purpose of primarily benefiting the website owner and not helping users. This practice is typically focused on creating large amounts of unoriginal content that provides little to no value for website visitors compared to other similar pages on the web, no matter how it's created.

規模化內容濫用＝大量產頁主要為站主獲利而非幫助使用者；典型是大量無原創性、相較同題其他頁面幾無價值的內容，不論怎麼產生。

Google — Search Quality Rater Guidelines §4.6.5（2025-09-11 版）｜官方 PDF｜抓取 2026-08-10

Using automated tools (generative AI or otherwise) as a low-effort way to produce many pages that add little-to-no value for website visitors as compared to other pages on the web on the same topic.

用自動化工具（生成式 AI 或其他）以低努力方式量產頁面、相較同題其他頁面幾無附加價值——是規模化內容濫用的例示。

Google — Search Quality Rater Guidelines §4.6.5（2025-09-11 版）｜官方 PDF｜抓取 2026-08-10

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.

規模化產出且無原創或附加價值的頁面一律評 Lowest（最低），不論產製方式；即使不確定是否用了生成式 AI，只要看過數頁後強烈懷疑是規模化濫用，仍應評 Lowest。

Google — Search Quality Rater Guidelines §4.6.5（2025-09-11 版）｜官方 PDF｜抓取 2026-08-10

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. Such pages should be rated Lowest, even if the page assigns credit for the content to another source.

頁面主內容（文字、圖片、音訊、影片等）全部或幾乎全部是低努力、低原創、低附加價值的複製、改寫、內嵌、自動或 AI 生成、轉貼時，評 Lowest——即使有標注出處。

Google — Search Quality Rater Guidelines §4.6.6（2025-09-11 版）｜官方 PDF｜抓取 2026-08-10

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. However, it's also possible to use Generative AI tools to create Lowest quality content with little to no effort, little to no originality, and little to no added value for website visitors.

單憑「用了生成式 AI」不決定努力程度或頁面品質評分；AI 工具既可做高品質創作，也可低努力產出 Lowest 品質內容。

Google — Search Quality Rater Guidelines §4.6.6（2025-09-11 版）｜官方 PDF｜抓取 2026-08-10

Content that is paraphrased from a single source or multiple sources. Content from a single page or from many websites can be summarized, reworded or paraphrased by people or generative AI tools. Paraphrasing may be valuable, for example when an expert paraphrases the contents of a government policy in easy-to-understand language.

改寫（人或生成式 AI 都可能做）可以有價值——例如專家把政府政策改寫成易懂語言。

Google — Search Quality Rater Guidelines §4.6.6（2025-09-11 版）｜官方 PDF｜抓取 2026-08-10

Important: Copying, paraphrasing, embedding or reposting content does not automatically make a page Lowest quality. However, the Lowest rating is required when all or almost all of the MC on the page (including text, images, audio, videos, etc) is copied, paraphrased, embedded, or reposted with little to no effort, little to no originality, and little to no added value for website visitors.

複製、改寫、內嵌、轉貼本身不會自動變 Lowest；但當主內容「全部或幾乎全部」如此、且低努力低原創低附加價值時，必須評 Lowest。

Google — Search Quality Rater Guidelines §4.6.6（2025-09-11 版）｜官方 PDF｜抓取 2026-08-10

Released the March 2025 core update. The rollout may take up to 2 weeks to complete.

發布 2025 年 3 月核心更新，推出最長需兩週完成。

Google Search Status Dashboard｜status.search.google.com/incidents/zpmwuSwifjDjfrVdaZUx｜抓取 2026-08-11

Released the June 2025 core update. The rollout may take up to 3 weeks to complete.

發布 2025 年 6 月核心更新，推出最長需三週完成。

Google Search Status Dashboard｜status.search.google.com/incidents/riq1AuqETW46NfBCe5NT｜抓取 2026-08-11

Released the August 2025 spam update, which applies globally and to all languages. The rollout may take a few days to complete.

發布 2025 年 8 月垃圾更新，適用全球所有語言。

Google Search Status Dashboard｜status.search.google.com/incidents/a7Aainy6E9rZsmfq82xt｜抓取 2026-08-11

Released the December 2025 core update. The rollout may take up to 3 weeks to complete.

發布 2025 年 12 月核心更新，推出最長需三週完成。

Google Search Status Dashboard｜status.search.google.com/incidents/DsirqJ1gpPRgVQeccPRv｜抓取 2026-08-11

Released the February 2026 Discover core update targeting English-language users in the US. The update aims to improve the quality of Discover overall and noted that standard guidance for core updates and Discover applies to this release. Future expansion to additional countries and languages planned.

發布 2026 年 2 月 Discover 核心更新，先針對美國英語用戶，旨在整體提升 Discover 品質，計畫再擴至其他國家語言。

Google Search Status Dashboard｜status.search.google.com/incidents/mYbNTqV1ytDc2fA8hUz4｜抓取 2026-08-11

Released the March 2026 spam update, which applies globally and to all languages. The rollout may take a few days to complete.

發布 2026 年 3 月垃圾更新，適用全球所有語言。

Google Search Status Dashboard｜status.search.google.com/incidents/VbnSXAH4SmEcxPtx4YSD｜抓取 2026-08-11

Released the March 2026 core update. A regular update designed to better surface relevant, satisfying content for searchers from all types of sites. The rollout may take up to 2 weeks to complete.

發布 2026 年 3 月核心更新：常規更新，旨在讓各類網站中相關、令人滿意的內容更容易浮現。

Google Search Status Dashboard｜status.search.google.com/incidents/7eTbAa2jWdToLkraZj5y｜抓取 2026-08-11

Released the May 2026 core update. A regular update designed to better surface relevant, satisfying content for searchers from all types of sites. The rollout may take up to 2 weeks to complete.

發布 2026 年 5 月核心更新：常規更新，措辭與 3 月核心更新相同。

Google Search Status Dashboard｜status.search.google.com/incidents/wdAXJk6LRRihEjpzEeWE｜抓取 2026-08-11

Released the June 2026 spam update, which applies globally and to all languages. This is a normal spam update, and it will roll out for all languages and locations. The rollout may take a few days to complete.

發布 2026 年 6 月垃圾更新：官方自述「這是一次正常的垃圾更新」，適用所有語言與地區。公告全文未提 AI、未提 slop、未列目標內容類型（見第 10 節與 7.0 注意事項）。

Google Search Status Dashboard｜status.search.google.com/incidents/YUX1peHev5a4fkxLDiUQ｜抓取 2026-08-11

This is a regular update designed to better surface relevant, satisfying content for searchers from all types of sites. We also continue our work to surface more content from creators through a series of improvements throughout this year.

官方定調：常規更新，並持續透過一系列改進讓更多創作者內容浮現。（Google Search Central 官方社群帳號聲明，無單獨永久連結，屬轉引；時間軸本體以 5.1 狀態頁為準）

Google Search Central（X／LinkedIn 官方帳號聲明，2025-03）｜對應狀態頁：status.search.google.com/incidents/zpmwuSwifjDjfrVdaZUx｜抓取 2026-08-11

There's nothing new or special that creators need to do for this update as long as they've been making satisfying content meant for people. For those that might not be ranking as well, we strongly encourage reading our Creating helpful, reliable, people-first content help page.

官方聲明：只要一直在做給人看的滿意內容，創作者不需為此更新做任何新動作；排名不佳者請讀 people-first 內容指引。（官方聲明之轉引；時間軸本體以 5.2 狀態頁為準）

Google（June 2025 core update 官方聲明）｜對應狀態頁：status.search.google.com/incidents/riq1AuqETW46NfBCe5NT｜抓取 2026-08-11

improve the experience by showing users more locally relevant content from websites based in their country, reducing sensational content and clickbait, and highlighting more in-depth, original, and timely content from sites with demonstrated expertise in a given area.

Discover 更新目標：多給用戶所在國的在地相關內容、減少聳動與釣魚內容、突出有實證專業網站的深度原創及時內容。

Google Search Central Blog — Discover core update（2026-02）｜developers.google.com/search/blog/2026/02/discover-core-update｜抓取 2026-08-11

Today, we're excited to announce the launch of new Search Generative AI performance reports in Search Console, including dedicated reports for Search and Discover, to help you understand your site's visibility within generative AI features on Search.

Search Console 推出生成式 AI 效能報表（Search 與 Discover 各有專屬報表），幫助網站主了解自己在生成式 AI 功能內的可見度。

Google Search Central Blog — Introducing Search Generative AI performance reports（2026-06-03）｜developers.google.com/search/blog/2026/06/gen-ai-performance-reports｜抓取 2026-08-11

The new Search Console reports are designed to give you dedicated views of your impressions within generative AI features on Search, such as AI Overviews and AI Mode, as well as generative AI features in Discover.

新報表提供網站在 AI Overviews、AI Mode 等生成式 AI 功能中的曝光專屬視圖。

Google Search Central Blog — 同文（2026-06-03）｜developers.google.com/search/blog/2026/06/gen-ai-performance-reports｜抓取 2026-08-11

This data is included in the overall performance report, where it will continue to be tracked to give site owners an overview of the overall visibility of their site in Google Search. Today, we are launching a separate view dedicated to visibility from generative AI features.

AI 功能數據仍計入整體效能報表；本次是另外新增一個生成式 AI 可見度的獨立視圖。

Google Search Central Blog — 同文（2026-06-03）｜developers.google.com/search/blog/2026/06/gen-ai-performance-reports｜抓取 2026-08-11

We are rolling these reports out to a subset of websites, allowing us to thoroughly test them and receive feedback before making them widely available.

報表先向一部分網站推出，充分測試並收集回饋後才會全面開放。（提醒：看不到報表不代表網站沒有在 AI 功能露出。）

Google Search Central Blog — 同文（2026-06-03）｜developers.google.com/search/blog/2026/06/gen-ai-performance-reports｜抓取 2026-08-11

Content creators and publishers use many channels beyond their own websites to reach their audiences. As people gravitate toward firsthand perspectives and different content formats, we want to make it easier for site owners and creators—even those without their own website—to get a consolidated view of how all of their content is getting discovered on Search.

因應使用者偏好第一手觀點與多元格式，Google 要讓創作者（即使沒有自己的網站）能整合看到其所有內容在搜尋上的被發現情形。

Google Search Central Blog — Search Console social & video platforms（2026-07-07）｜developers.google.com/search/blog/2026/07/search-console-social-video-platforms｜抓取 2026-08-11

Following an earlier experiment, we're thrilled to introduce platform properties, a new Search Console property type to help site owners and creators understand how their social and video posts perform on Google Search and Discover.

推出「platform properties」：Search Console 新資產類型，用於了解社群與影音貼文在 Google 搜尋與 Discover 的表現。

Google Search Central Blog — 同文（2026-07-07）｜developers.google.com/search/blog/2026/07/search-console-social-video-platforms｜抓取 2026-08-11

Now, you can track which search terms lead people to your Instagram, TikTok, X, and YouTube content on Search, and see exactly how your audience is interacting with your posts.

現在可以追蹤哪些搜尋字詞把人帶到你的 Instagram、TikTok、X、YouTube 內容，並看到受眾如何互動。

Google Search Central Blog — 同文（2026-07-07）｜developers.google.com/search/blog/2026/07/search-console-social-video-platforms｜抓取 2026-08-11

Earlier this month, we announced platform properties for Search Console, allowing you to track how your social and video posts on Instagram, TikTok, X, and YouTube perform on Google Search, Discover, and Google News. Today, platform properties are globally available to everyone.

平台 Properties 自 2026-07-29 起全球開放給所有人。

Google Search Central Blog — Platform properties roll out globally（2026-07-29）｜developers.google.com/search/blog/2026/07/platform-properties-social-video-guide｜抓取 2026-08-11

Today, we're also publishing a new guide on how to analyze your social and video content performance. Whether you're a content creator, social media manager, or SEO professional, this guide will show you what insights are available from Search Console and how to analyze your performance.

同日發布社群與影音內容成效分析新指南，說明 Search Console 提供哪些洞察與分析方法。

Google Search Central Blog — 同文（2026-07-29）｜developers.google.com/search/blog/2026/07/platform-properties-social-video-guide｜抓取 2026-08-11

Overall, total organic click volume from Google Search to websites has been relatively stable year-over-year.

Google（Liz Reid）：從 Google 搜尋導向網站的總自然點擊量，年比相對穩定。

Google — AI in Search: driving more queries and higher quality clicks（2025-08-06）｜blog.google/.../ai-search-driving-more-queries-higher-quality-clicks/｜抓取 2026-08-11

Additionally, average click quality has increased and we're actually sending slightly more quality clicks to websites than a year ago.

Google（Liz Reid）：平均點擊品質提升，送往網站的優質點擊比一年前略多。

Google — 同文（2025-08-06）｜blog.google/.../ai-search-driving-more-queries-higher-quality-clicks/｜抓取 2026-08-11

For some questions where people are looking for a quick answer, like 'when is the next full moon,' people may be satisfied with the initial response and not click further.

Google（Liz Reid）：找快速答案的查詢（如「下次滿月是何時」），用戶可能滿足於初始回應而不再點擊。

Google — 同文（2025-08-06）｜blog.google/.../ai-search-driving-more-queries-higher-quality-clicks/｜抓取 2026-08-11

We've seen that when people click to a website from search results pages with AI Overviews, these clicks are higher quality, where users are more likely to spend more time on the site.

Google（John Mueller）：從帶 AI Overviews 的結果頁點進網站的點擊品質更高，用戶更可能停留更久。

Google Search Central Blog — Succeeding in AI Search（2025-05-21）｜developers.google.com/search/blog/2025/05/succeeding-in-ai-search｜抓取 2026-08-11

Focus on your visitors and provide them with unique, satisfying content.

Google（John Mueller）：聚焦你的訪客，提供獨特、令人滿意的內容。

Google Search Central Blog — 同文（2025-05-21）｜developers.google.com/search/blog/2025/05/succeeding-in-ai-search｜抓取 2026-08-11

Focus on making unique, non-commodity content that visitors from Search and your own readers will find helpful and satisfying.

Google（John Mueller）：專注做獨特、非同質化（non-commodity）、讓搜尋訪客與自家讀者都覺得有用且滿意的內容。

Google Search Central Blog — 同文（2025-05-21）｜developers.google.com/search/blog/2025/05/succeeding-in-ai-search｜抓取 2026-08-11

Users who encountered an AI summary clicked on a traditional search result link in 8% of all visits. Those who did not encounter an AI summary clicked on a search result nearly twice as often (15% of visits).

遇到 AI 摘要的用戶只在 8% 的造訪中點了傳統搜尋結果連結；沒遇到 AI 摘要者點擊率近兩倍（15%）。

Pew Research Center（2025-07-22）｜pewresearch.org/short-reads/2025/07/22/...｜抓取 2026-08-11

Google users who encountered an AI summary also rarely clicked on a link in the summary itself. This occurred in just 1% of all visits to pages with such a summary.

遇到 AI 摘要的用戶也極少點摘要裡的連結——只發生在 1% 的造訪中。

Pew Research Center（2025-07-22）｜pewresearch.org/short-reads/2025/07/22/...｜抓取 2026-08-11

Google users are more likely to end their browsing session entirely after visiting a search page with an AI summary than on pages without a summary. This happened on 26% of pages with an AI summary, compared with 16% of pages with only traditional search results.

造訪帶 AI 摘要的搜尋頁後直接結束整個瀏覽的比率是 26%，只有傳統結果的頁面則是 16%。

Pew Research Center（2025-07-22）｜pewresearch.org/short-reads/2025/07/22/...｜抓取 2026-08-11

Around two-thirds of all searches resulted in one of these actions.

全部搜尋中約三分之二以上述行為之一（不點擊／結束瀏覽等）收場。

Pew Research Center（2025-07-22）｜pewresearch.org/short-reads/2025/07/22/...｜抓取 2026-08-11

We found that the presence of an AI Overview now correlates with a 58% lower average clickthrough rate for the top-ranking page.

Ahrefs：AI Overview 的存在，與排名第一頁面的平均點閱率低 58% 相關（原文為 correlates with，非因果證明）。

Ahrefs Blog（2026-02-04）｜ahrefs.com/blog/ai-overviews-reduce-clicks-update/｜抓取 2026-08-11

In December 2023, the average position one CTR for AI Overview keywords was 0.073. In December 2025, this has dropped to 0.016.

AI Overview 關鍵字的第一名平均 CTR：2023 年 12 月 0.073 → 2025 年 12 月 0.016。

Ahrefs Blog（2026-02-04）｜ahrefs.com/blog/ai-overviews-reduce-clicks-update/｜抓取 2026-08-11

Overall, Similarweb estimates that AI Platforms generated over 1.13 billion referral visits in June 2025.

Similarweb 估算：2025 年 6 月 AI 平台產生超過 11.3 億次引薦造訪（estimated metrics，clickstream 類估算）。

Similarweb Blog（2025-07-29）｜similarweb.com/blog/insights/ai-news/ai-referral-traffic-winners/｜抓取 2026-08-11

In June, AI referrals were up 357% from June 2024.

2025 年 6 月的 AI 引薦流量比 2024 年 6 月成長 357%。

Similarweb Blog（2025-07-29）｜similarweb.com/blog/insights/ai-news/ai-referral-traffic-winners/｜抓取 2026-08-11

June's referrals to news and media websites were up 770% from June 2024.

導向新聞媒體網站的 AI 引薦流量年增 770%。

Similarweb Blog（2025-07-29）｜similarweb.com/blog/insights/ai-news/ai-referral-traffic-winners/｜抓取 2026-08-11

In the first four months of 2026, a whopping 68.01% of Google searches ended without a click.

2026 年前四個月，68.01% 的 Google 搜尋以零點擊收場（資料源為 Similarweb desktop/mobile panel）。

SparkToro Blog（2026-06-08）｜sparktoro.com/blog/in-2026-less-than-one-third-of-google-searches-still-send-a-click/｜抓取 2026-08-11

In 2024, US zero click searches on Google stood at 60.45%.

2024 年美國 Google 零點擊搜尋為 60.45%（對照 7.16 的 2026 數字）。

SparkToro Blog（2026-06-08）｜sparktoro.com/blog/in-2026-less-than-one-third-of-google-searches-still-send-a-click/｜抓取 2026-08-11

The decline from June 2025 to May 2026 is 8 percentage points, a ~22% drop in the traffic share Google sends to these tens of thousands of websites.

2025 年 6 月至 2026 年 5 月，Google 送往上萬網站的流量占比下滑 8 個百分點，約 22% 的降幅。

SparkToro Blog（2026-06-08）｜sparktoro.com/blog/in-2026-less-than-one-third-of-google-searches-still-send-a-click/｜抓取 2026-08-11

Google and their expansion of AI Overviews continues to keep web traffic captive in the Google search experience and migrate search to Gemini.

Chegg CEO Nathan Schultz：Google 擴張 AI Overviews，持續把網路流量困在 Google 搜尋體驗內，並把搜尋遷往 Gemini。

Chegg 2025 Q1 財報／法說（2025-05-12）｜investor.chegg.com/.../Chegg-Reports-2025-First-Quarter-Earnings/｜抓取 2026-08-11

Looking ahead, industry challenges continue to cause a notable decline in traffic and subscriber acquisitions.

Chegg CFO 準備稿：展望未來，行業挑戰持續造成流量與訂閱獲取顯著下滑。

Chegg 2025 Q1 財報／法說（2025-05-12）｜investor.chegg.com/.../Chegg-Reports-2025-First-Quarter-Earnings/｜抓取 2026-08-11

Total Net Revenues of $121.4 million, a decrease of 30% year-over-year

Chegg 總淨營收 1.214 億美元，年減 30%。（注意：財報未把此下滑單一歸因於 AI Overviews，引用時不可作此歸因。）

Chegg 2025 Q1 財報（2025-05-12）｜investor.chegg.com/.../Chegg-Reports-2025-First-Quarter-Earnings/｜抓取 2026-08-11

the site was experiencing a large drop in clicks from search results featuring an AI summary, with clickthrough rates dropping by 56.1% on the desktop site and 48.2% on mobiles.

MailOnline（Carly Steven，The Guardian 轉述）：帶 AI 摘要的搜尋結果讓點閱率在桌機掉 56.1%、行動掉 48.2%。

The Guardian（2025-07-24，具名公開發言之轉述）｜theguardian.com/technology/2025/jul/24/...｜抓取 2026-08-11

a site previously ranked first in a search result could lose about 79% of its traffic for that query if results were delivered below an AI overview.

Authoritas 研究（The Guardian 轉述）：原本排第一的網站，若結果被壓在 AI overview 下方，該查詢流量可能損失約 79%。

The Guardian（2025-07-24，行業研究轉述）｜theguardian.com/technology/2025/jul/24/...｜抓取 2026-08-11

Business Insider CEO Barbara Peng laid off about 21% of her staff in May, citing 'extreme traffic declines beyond our control' as the reason for the layoffs.

Business Insider CEO Barbara Peng 於 5 月裁員約 21%，理由是「超出我們控制的極端流量下滑」（Euronews 轉述）。

Euronews（2025-08-03，具名公開發言之轉述）｜euronews.com/next/2025/08/03/...｜抓取 2026-08-11

Google is changing from a search engine to an answer engine.

The Atlantic 執行長 Nicholas Thompson：Google 正從搜尋引擎變成答案引擎（Euronews 轉述）。

Euronews（2025-08-03，具名公開發言之轉述）｜euronews.com/next/2025/08/03/...｜抓取 2026-08-11

Generative engine optimization (GEO) is the practice of optimizing your presence and content to appear in responses generated by AI-powered search systems such as ChatGPT, Google, Perplexity, Claude, and others.

GEO＝優化你的存在與內容，使其出現在 ChatGPT、Google、Perplexity、Claude 等 AI 搜尋系統生成的回應中。（GEO 是行業用語，非 Google 官方術語。）

Semrush Blog（2026-04-16）｜semrush.com/blog/generative-engine-optimization/｜抓取 2026-08-11

The key difference between SEO and GEO is that you aren't competing to rank at the top of search results in GEO—you're competing to be part of the final output.

SEO 與 GEO 的關鍵差異：GEO 不是搶結果頁排名，而是搶成為最終輸出的一部分。

Semrush Blog（2026-04-16）｜semrush.com/blog/generative-engine-optimization/｜抓取 2026-08-11

We're making a minor update to our 'repetitious content' policy to better clarify this includes content that is repetitive or mass-produced. We are also renaming this policy from 'repetitious content' to 'inauthentic content.'

2025-07-15 起，YouTube 把「重複性內容」政策更名為「不真實內容（inauthentic content）」，明確涵蓋重複或量產內容。

YouTube — Channel monetization policies 更新公告（2025-07-15）｜support.google.com/youtube/answer/1311392｜抓取 2026-08-10

AI-generated content made with generic or unoriginal templates giving the impression of mass production

用泛用或非原創模板做出的、給人量產印象的 AI 生成內容——列為不真實內容違規例示。

YouTube — Channel monetization policies（inauthentic content 例示）｜support.google.com/youtube/answer/1311392｜抓取 2026-08-10

the substance of each video should be materially varied and deliver creative, educational, or other value.

每支影片的實質內容應有實質差異，並提供創意、教育或其他價值。

YouTube — Channel monetization policies（acceptable content 原則）｜support.google.com/youtube/answer/1311392｜抓取 2026-08-10

Automated or synthetic mass-production: Using automated tools or AI to churn out high volumes of similar content with minimal changes. While testing out new creation tools or posting a few variations of a video is ok, we don't allow use of these tools to flood our platform with repetitive content. This includes coordinated mass-production and technical manipulation to bypass filters or trick viewers.

用自動化工具或 AI 以極小變化大量產出相似內容即違規；測試新工具或發幾個變體可以，但不准用這些工具以重複內容灌爆平台，包括協同量產與繞過過濾器的技術操縱。

YouTube — Spam, deceptive practices & scams policies｜support.google.com/youtube/answer/2801973｜抓取 2026-08-10

To help keep viewers informed about the content they're viewing, we require creators to disclose when they use AI to meaningfully alter or generate photorealistic content.

創作者用 AI 實質改動或生成逼真（photorealistic）內容時，必須揭露。

YouTube — Disclosing use of altered or synthetic content｜support.google.com/youtube/answer/14328491｜抓取 2026-08-10

Disclosing AI content won't limit a video's audience or impact its eligibility to earn money.

揭露 AI 內容不會限制影片觸及，也不影響營利資格。（反過來：持續不揭露者可能被強制標示、移除內容或停權 YPP。）

YouTube — Disclosing use of altered or synthetic content｜support.google.com/youtube/answer/14328491｜抓取 2026-08-10

We do not allow content that is designed to deceive, mislead, or overwhelm users in order to artificially increase viewership. This content detracts from people's ability to engage authentically on our platforms and can threaten the security, stability and usability of our services.

不允許為人工灌量而設計來欺騙、誤導或轟炸使用者的內容。

Meta — Community Standards: Spam（Transparency Center）｜transparency.meta.com/policies/community-standards/spam/｜抓取 2026-08-10

Creators who are not adding substantial creative value will see their content deprioritized, and they may see their account deemed non-recommendable and demonetized, if they continue to post primarily unoriginal content.

不添加實質創意價值的創作者，內容會被降權；若持續主要發非原創內容，帳號可能被判定不可推薦並取消營利。

Meta Newsroom — Rewarding original creators on Facebook（2026-03）｜about.fb.com/news/2026/03/rewarding-original-creators-on-facebook/｜抓取 2026-08-10

Third-party content that is posted without substantial changes is considered unoriginal and will receive reduced distribution. Posting unoriginal content violates Meta's terms and Partner Monetisation Policies, and may also lead to additional enforcement actions such as limited or loss of recommendations and monetisation abilities.

未做實質改動就轉發的第三方內容視為非原創，會被降發佈；發非原創內容違反 Meta 條款與夥伴營利政策，並可能失去推薦與營利能力。

Meta — Original Content Guidelines（Facebook Business Help）｜en-gb.facebook.com/business/help/262834734651607｜抓取 2026-08-10

Content creators and publishers can only monetize content that they created or were involved in the creation of, or that directly features the creator or publisher. Content that is unoriginal or reproduced without making meaningful enhancements (commentary, parody, creative editing, etc.) cannot be monetized.

只能對自己創作、參與創作或直接以本人為主角的內容營利；未做有意義強化（評論、戲仿、創意剪輯等）的非原創或翻製內容不可營利。

Instagram — Content Monetization Policies｜help.instagram.com/2635536099905516｜抓取 2026-08-10

we will label images that users post to Facebook, Instagram and Threads when we can detect industry standard indicators that they are AI-generated.

當能偵測到業界標準的 AI 生成指標時，Meta 會對 Facebook、Instagram、Threads 上的用戶圖片加標。

Meta Newsroom — Labeling AI-generated images（2024-02）｜about.fb.com/news/2024/02/labeling-ai-generated-images-on-facebook-instagram-and-threads/｜抓取 2026-08-10

Meta uses industry-standard detection methods, such as C2PA to identify when ad content has been created or edited using third-party generative AI tools. When we detect this metadata, we label the content accordingly.

Meta 用 C2PA 等業界標準方法偵測廣告內容是否經第三方生成式 AI 產製或編輯，偵測到中繼資料即據以標示。

Meta — AI Info on Ads（Facebook Business Help）｜facebook.com/business/help/1010479435004531｜抓取 2026-08-10

We welcome creativity, including when it comes from new digital tools like generative artificial intelligence (AI). But generative AI and editing can blur the line between fact and fiction. To help keep content on TikTok trustworthy and provide people with important context about what they are viewing, we require creators to label AI-generated or significantly edited content that shows realistic-looking scenes or people.

歡迎 AI 創意，但生成式 AI 會模糊真實與虛構的界線；凡呈現逼真場景或人物的 AI 生成或重度編輯內容，創作者必須標示。

TikTok — Community Guidelines: Integrity and Authenticity（2025-08-14 發布／2025-09-13 生效）｜tiktok.com/community-guidelines/en/integrity-authenticity｜抓取 2026-08-10

Unlabeled content may be removed, restricted, or labeled by our team, depending on the harm it could cause.

未標示的內容可能依其潛在危害被移除、限制或由官方代為標示。

TikTok — Community Guidelines: AIGC（2025-08-14 版）｜tiktok.com/community-guidelines/en/integrity-authenticity｜抓取 2026-08-10

Content that violates our Community Guidelines is never allowed on TikTok, regardless of whether it was created or altered with AI

違反社群守則的內容一律不允許，不論是否用 AI 產製或改動。

TikTok — Supporting responsible, transparent AI-generated content｜tiktok.com/transparency/en/supporting-responsible-transparent-ai-generated-content/｜抓取 2026-08-10

Your account info or video content is copied from others or has minimal original input or edits

帳號資訊或影片內容抄自他人、或原創投入與編輯極少——即失去創作者獎勵計畫（Creator Rewards）資格。

TikTok — Creator Academy: Creator Rewards Program Eligibility｜tiktok.com/creator-academy/article/eligibility｜抓取 2026-08-10

Reproduced or unoriginal content that is imported or uploaded without any new or creative edits, such as content with someone else's visible watermark or superimposed logo

無任何新增或創意編輯就搬運上傳的翻製／非原創內容（例如帶著他人浮水印或疊加 logo 的內容），不具 For You Feed 推薦資格。

TikTok — For You Feed Eligibility Standards（2024-04-17 發布）｜tiktok.com/safety/en/policies-and-engagement/fyf-standards｜抓取 2026-08-10

We also use a cross-industry technology called C2PA Content Credentials, which embeds metadata into content that lets us know when something is AI-generated

TikTok 使用跨業界的 C2PA Content Credentials 技術，靠內容內嵌的中繼資料辨識 AI 生成內容並自動標示。

TikTok Newsroom — More ways to spot, shape and understand AI content｜newsroom.tiktok.com/more-ways-to-spot-shape-and-understand-ai-content｜抓取 2026-08-10

You may not engage in inauthentic activity that undermines the integrity of X.

不得從事破壞 X 完整性的不真實活動。

X — Authenticity（版本 April 2025；ego-browser 真瀏覽器抓取）｜help.x.com/en/rules-and-policies/authenticity｜抓取 2026-08-10

Fake personas: You may not use manufactured identities to engage in disruptive or deceptive behavior. This may include using stock, stolen or AI-generated profile photos, copied or stolen profile bios, and/or misleading profile information for the purpose of deceiving others.

不得用捏造身分從事擾亂或欺騙行為，包括為騙人而使用素材庫、盜用或 AI 生成的頭像、抄襲的簡介或誤導性個資。

X — Authenticity（版本 April 2025）｜help.x.com/en/rules-and-policies/authenticity｜抓取 2026-08-10

Content Spam: You may not share or post content in a bulk, duplicative, irrelevant or unsolicited manner that disrupts people's experience.

不得以大量、重複、無關或未經請求的方式發布內容干擾他人體驗。

X — Authenticity（版本 April 2025）｜help.x.com/en/rules-and-policies/authenticity｜抓取 2026-08-10

Synthetic and Manipulated Media: You may not share inauthentic media, including, manipulated, or out-of-context media that may result in widespread confusion on public issues, impact public safety, or cause serious harm ("misleading media"). We may use our own technology or receive reports through partnerships with third parties in order to determine if media have been manipulated or presented out of context. In situations where we are unable to reliably determine if it is misleading media, we may not take action.

不得分享可能在公共議題上造成廣泛混淆、影響公共安全或造成嚴重傷害的不真實媒體（操縱或斷章取義的媒體）；X 會用自家技術或第三方合作判定，無法可靠判定時可能不採取行動。

X — Authenticity（版本 April 2025）｜help.x.com/en/rules-and-policies/authenticity｜抓取 2026-08-10

Videos depicting armed conflict: For any AI generated content depicting armed conflict, you must add a clear disclosure that it is AI generated. Additionally, you must also label realistic video game depictions of armed conflict as a synthetic video. You may not attempt to mislead people about the origin of any depictions of armed conflict, including by claiming they are from a different armed conflict or a different time period.

凡描繪武裝衝突的 AI 生成內容，必須清楚揭露為 AI 生成；逼真的遊戲畫面也須標示為合成影片；不得誤導衝突影像的來源。這是 X 營利標準中唯一明文的 AI 揭露義務條。

X — Creator Monetization Standards｜help.x.com/en/rules-and-policies/content-monetization-standards｜抓取 2026-08-10

in the past 12 months alone, a period marked by the explosion of generative AI tools, we've removed over 75 million spammy tracks from Spotify.

在生成式 AI 工具爆發的過去 12 個月內，Spotify 移除了逾 7,500 萬首垃圾曲目。

Spotify Newsroom — Spotify strengthens AI protections（2025-09-25）｜newsroom.spotify.com/2025-09-25/spotify-strengthens-ai-protections/｜抓取 2026-08-10

This fall, we'll roll out a new music spam filter—a system that will identify uploaders and tracks engaging in these tactics, tag them, and stop recommending them.

Spotify 推出音樂垃圾過濾器：辨識採用垃圾手法的上傳者與曲目，加標並停止推薦。

Spotify Newsroom — 同文（2025-09-25）｜newsroom.spotify.com/2025-09-25/spotify-strengthens-ai-protections/｜抓取 2026-08-10

We're helping develop and will support the new industry standard for AI disclosures in music credits, developed through DDEX.

Spotify 協助開發並將支持透過 DDEX 制定的音樂製作名單 AI 揭露新業界標準。

Spotify Newsroom — 同文（2025-09-25）｜newsroom.spotify.com/2025-09-25/spotify-strengthens-ai-protections/｜抓取 2026-08-10

Vocal impersonation is only allowed in music on Spotify when the impersonated artist has authorized the usage.

聲音模仿只有在被模仿的藝人授權時才被允許。

Spotify Newsroom — 同文（2025-09-25）｜newsroom.spotify.com/2025-09-25/spotify-strengthens-ai-protections/｜抓取 2026-08-10

Deezer is receiving nearly 75,000 AI generated tracks every day – accounting for more than 44% of the total daily delivery.

Deezer 每天收到近 75,000 首 AI 生成曲目，占每日總交付量逾 44%（2026 年 4 月數字）。

Deezer Newsroom（2026-04）｜newsroom-deezer.com/2026/04/ai-generated-tracks-represent-44-of-new-uploaded-music/｜抓取 2026-08-10

Up to 85% of streams on these tracks were identified as fraudulent and excluded from royalty calculations

這些（全 AI 生成）曲目高達 85% 的串流被判定為造假，並被剔除於版稅計算之外。

Deezer Newsroom — Deezer Detector（2026-06）｜newsroom-deezer.com/2026/06/check-ai-generated-music-in-playlists-with-deezer-detector/｜抓取 2026-08-10

Tracks identified and detected as AI-generated are automatically excluded from algorithmic recommendations and editorial playlists

被偵測為 AI 生成的曲目自動排除於演算法推薦與編輯歌單之外（但仍留在平台上並清楚標示 AI 標籤）。

Deezer Newsroom — Deezer Detector（2026-06）｜newsroom-deezer.com/2026/06/check-ai-generated-music-in-playlists-with-deezer-detector/｜抓取 2026-08-10

Text, images, or translations created by an AI-based tool. If you used an AI-based tool to create the actual content (whether text, images, or translations), it is considered 'AI-generated,' even if you applied substantial edits afterwards.

凡由 AI 工具「創建」的文字、圖片或翻譯即屬 AI-generated——即使事後做了大量編輯也算。

Amazon KDP — Content Guidelines｜kdp.amazon.com/en_US/help/topic/G200672390｜抓取 2026-08-10

If you created the content yourself, and used AI-based tools to edit, refine, error-check, or otherwise improve that content (whether text or images), then it is considered 'AI-assisted' and not 'AI-generated.'

內容自己創作、只用 AI 工具編輯、打磨、查錯或改進——屬 AI-assisted，不算 AI-generated。

Amazon KDP — Content Guidelines｜kdp.amazon.com/en_US/help/topic/G200672390｜抓取 2026-08-10

We require you to inform us of AI-generated content (text, images, or translations) when you publish a new book or make edits to and republish an existing book through KDP. You are not required to disclose AI-assisted content.

出版新書或改版重出時，AI-generated 內容必須向 Amazon 申報；AI-assisted 內容不需揭露。

Amazon KDP — Content Guidelines｜kdp.amazon.com/en_US/help/topic/G200672390｜抓取 2026-08-10

writing where the majority of the content has been created by an AI-writing program with little or no edits, improvements, fact-checking, or changes

AI-generated 寫作＝多數內容由 AI 寫作程式產生、幾乎未經編輯、改進、查核或修改。

Medium — AI content policy｜help.medium.com/.../Artificial-Intelligence-AI-content-policy｜抓取 2026-08-10

"Network Only" distribution—limited to "the author's direct network of followers and subscribers only."

未揭露的 AI 內容只獲「Network Only」發佈——僅限作者的直接跟隨者與訂閱者可見。

Medium — AI content policy｜help.medium.com/.../Artificial-Intelligence-AI-content-policy｜抓取 2026-08-10

AI-generated writing (disclosed as such or not) is not allowed to be paywalled

AI 生成的寫作（不論有無揭露）一律不得放進付費牆。

Medium — AI content policy｜help.medium.com/.../Artificial-Intelligence-AI-content-policy｜抓取 2026-08-10

The use of large language models (LLMs; the "engine" behind AI chatbots, such as ChatGPT) on Wikipedia presents systemic risks to maintaining the content standards required by the core content policies, specifically through the introduction of "hallucinated" statements, unsourced or unverifiable content, and algorithmic bias.

在維基百科使用 LLM 對核心內容政策構成系統性風險：幻覺陳述、無來源或不可查證內容、演算法偏誤。

Wikipedia — Wikipedia:Large language models（2024-2025）｜en.wikipedia.org/wiki/Wikipedia:Large_language_models｜抓取 2026-08-10

For this reason, using LLMs to generate or rewrite article content is prohibited, except for translation and for basic copyediting of one's own work.

禁止用 LLM 產生或重寫條目內容，僅翻譯與自己作品的基本文字校訂除外。

Wikipedia — Wikipedia:Large language models（2024-2025）｜en.wikipedia.org/wiki/Wikipedia:Large_language_models｜抓取 2026-08-10

LLM-created works are not reliable sources. Unless their outputs were published by reliable outlets with rigorous oversight, and unless it can be verified that the content were evaluated for accuracy by the publisher, they should not be cited.

LLM 產物不是可靠來源；除非其輸出由有嚴格把關的可靠媒體發表且可驗證經過準確性審核，否則不應被引用。

Wikipedia — Wikipedia:Large language models（2024-2025）｜en.wikipedia.org/wiki/Wikipedia:Large_language_models｜抓取 2026-08-10

automated excessive bulk activity and coordinated inauthentic activity, such as spamming; cryptocurrency mining

GitHub 可接受使用政策禁止：自動化過量批次活動與協同不真實活動（如濫發垃圾）、挖礦。

GitHub — Acceptable Use Policies｜docs.github.com/en/site-policy/acceptable-use-policies/github-acceptable-use-policies｜抓取 2026-08-10

inauthentic interactions, such as fake accounts and automated inauthentic activity

禁止不真實互動：假帳號、自動化不真實活動。

GitHub — Acceptable Use Policies｜docs.github.com/en/site-policy/acceptable-use-policies/github-acceptable-use-policies｜抓取 2026-08-10

while the answers which ChatGPT produces have a high rate of being incorrect, they typically look like they might be good and the answers are very easy to produce

ChatGPT 產生的答案錯誤率高，卻通常看起來像對的，而且產生成本極低——這是 Stack Overflow 2022 年 12 月禁令的原始理由（政策後續有調整，以現行頁為準）。

Stack Overflow — Help Center: gen-ai policy（2022-12 起）｜stackoverflow.com/help/gen-ai-policy｜抓取 2026-08-10

deceit, fraud, scams, spam, or impersonation

OpenAI 使用政策禁止把服務用於：欺瞞、詐欺、詐騙、垃圾訊息或冒充。

OpenAI — Usage Policies｜openai.com/policies/usage-policies｜抓取 2026-08-10

Promote or facilitate the generation or distribution of spam

Anthropic 使用政策禁止：促成或協助垃圾內容的產生與散布。

Anthropic — Usage Policy（AUP）｜anthropic.com/legal/aup｜抓取 2026-08-10

Impersonate real entities or create fake personas to falsely attribute content or mislead others about its origin without consent or legal right

禁止冒充真實實體或建立假人設，以不實歸屬內容或誤導他人其來源。

Anthropic — Usage Policy（AUP）｜anthropic.com/legal/aup｜抓取 2026-08-10

Generate deceptive or misleading digital content such as fake reviews, comments, or media

禁止產生欺騙或誤導性的數位內容，如假評論、假留言、假媒體。

Anthropic — Usage Policy（AUP）｜anthropic.com/legal/aup｜抓取 2026-08-10

Misrepresenting the provenance of generated content by claiming it was created solely by a human, in order to deceive

Google 生成式 AI 禁用政策：禁止為欺騙而謊稱生成內容純由人類創作。

Google — Generative AI Prohibited Use Policy｜policies.google.com/terms/generative-ai/use-policy｜抓取 2026-08-10

Impersonating an individual (living or dead) without explicit disclosure, in order to deceive

禁止為欺騙而未明確揭露地冒充在世或已故的個人。

Google — Generative AI Prohibited Use Policy｜policies.google.com/terms/generative-ai/use-policy｜抓取 2026-08-10

You may not intentionally mislead recipients of the Assets about their nature or source

Midjourney 服務條款：不得刻意誤導接收者關於產出物的性質或來源。

Midjourney — Terms of Service｜docs.midjourney.com/docs/terms-of-service｜抓取 2026-08-10

To represent or imply that an Output generated from our Services was human-generated, when it was not

Stability AI 服務條款禁止：把服務產出表述或暗示為人工產出。

Stability AI — Terms of Service｜stability.ai/terms-of-service｜抓取 2026-08-10

Representing that Llama outputs are human-generated when they are not

Meta Llama 可接受使用政策禁止：把 Llama 輸出表述為人類產出。

Meta — Llama Acceptable Use Policy｜llama.com/llama4/use-policy/｜抓取 2026-08-10

Generating or facilitating false online engagement, including fake reviews

禁止產生或協助不實的線上互動，包括假評論。

Meta — Llama Acceptable Use Policy｜llama.com/llama4/use-policy/｜抓取 2026-08-10

Digital media created algorithmically using an Artificial Intelligence model trained on captured content

trainedAlgorithmicMedia＝以「用實拍內容訓練的 AI 模型」演算法產生的數位媒體。此即 Google Merchant Center 要求 AI 圖片必帶的標記值。

IPTC — Digital Source Type NewsCodes（建立 2022-06-14／修訂 2024-10-23）｜cv.iptc.org/newscodes/digitalsourcetype/｜抓取 2026-08-10

Mix or composite of several elements, at least one of which is Generative AI

compositeSynthetic＝多元素合成，其中至少一個元素是生成式 AI。

IPTC — Digital Source Type NewsCodes（2022-06-14）｜cv.iptc.org/newscodes/digitalsourcetype/｜抓取 2026-08-10

Media created purely by an algorithm not based on any sampled training data, e.g. an image created by software using a mathematical formula

algorithmicMedia＝不基於任何取樣訓練資料、純由演算法產生的媒體（如數學公式生成的圖）——與 AI 訓練模型產物是不同碼。

IPTC — Digital Source Type NewsCodes（2022-06-14）｜cv.iptc.org/newscodes/digitalsourcetype/｜抓取 2026-08-10

Augmentation, correction or enhancement using a Generative AI model, such as with inpainting or outpainting operations

「以生成式 AI 編輯」＝用生成式 AI 模型做增補、修正或強化（如 inpainting／outpainting）。

IPTC — Digital Source Type NewsCodes（建立 2023-07-26／修訂 2024-09-17）｜cv.iptc.org/newscodes/digitalsourcetype/｜抓取 2026-08-10

As part of an ongoing effort to enhance AI transparency and accountability, a new AI Disclosure assertion is introduced to provide a means for a Claim Generator to provide machine-readable AI transparency information. This assertion enables verifiable AI transparency, automated compliance verification, and trustworthy AI content at scale.

C2PA 2.4 引入 AI Disclosure assertion：提供機器可讀的 AI 透明資訊，支持可驗證的 AI 透明度、自動化合規驗證與規模化可信 AI 內容。

C2PA — Specification v2.4 §18.28.1（2026-04-23 發布）｜spec.c2pa.org/.../C2PA_Specification.html#_ai_disclosure｜抓取 2026-08-10

The AI Disclosure assertion shall have a label of `c2pa.ai-disclosure`.

AI Disclosure assertion 的標籤名為 c2pa.ai-disclosure。

C2PA — Specification v2.4 §18.28.2（2026-04-23 發布）｜spec.c2pa.org/.../C2PA_Specification.html#_ai_disclosure｜抓取 2026-08-10

It is important to identify the output as coming from a trained AI-ML model by using the appropriate digital source type. For a generative model this will be `trainedAlgorithmicMedia`, and for output that is not a media asset, the `c2pa.trainedAlgorithmicData` designation would be appropriate.

AI-ML 模型的輸出應以適當的 digital source type 標明：生成式模型用 trainedAlgorithmicMedia，非媒體資產的輸出用 c2pa.trainedAlgorithmicData。

C2PA — AI-ML Guidance v2.3 §9｜spec.c2pa.org/.../ai_ml.html#output-content-credential｜抓取 2026-08-10

The DDEX AI disclosure standard defines seven categories: (1) Vocals—AI used to generate or modify vocal performances; (2) Lyrics—lyrics written or co-written by AI; (3) Composition—musical composition (melody, harmony, chord progressions) generated by AI; (4) Production—AI used in the production process, including AI-generated instrumentals, beat creation, arrangement decisions; (5) Mixing—AI involved in the mixing process including automated level balancing and AI-driven effects processing; (6) Mastering—AI used for mastering via services like LANDR or CloudBounce; (7) Instrumentation—specific instrumental parts generated by AI.

DDEX AI 揭露標準拒絕「是／否 AI」二元選項，改為七類角色化揭露：人聲、歌詞、作曲、製作、混音、母帶、樂器聲部。此段為第三方對標準結構的整理；DDEX 正式標準入口見 ddex.net（ERN 4.3.2 + AI Extension, 2025）；Spotify 官方已宣布支持（見 8.26）。

DDEX AI Disclosure Extension（2025）之第三方整理｜undetectr.com/blog/ddex-ai-disclosure-standard-explained｜抓取 2026-08-10

Google News 於 2025 年開始過濾／排除 AI 或自動生成的發布商內容。

此說法常引用 Google News 說明頁 answer/15898024 作為依據。查證結果：該頁不含任何過濾自動生成發布商內容的政策文字。它宣布的是 Google News 自身架構轉為「自動生成的出版物頁面」，是 Google 自己的產品改版，不是對發布商的內容限制。

查證方法：2026-08-10 以 r.jina.ai 全文抓取該頁（48 行）逐字比對。頁面：support.google.com/news/publisher-center/answer/15898024（發布 2025-02-10／更新 2025-03-20）

This doesn't change what content is eligible to appear in Google News and other news surfaces. Content from publishers that adheres to our content policies is automatically eligible for consideration in Google News and across news surfaces.

官方原文：這（架構轉換）不改變什麼內容有資格出現在 Google News；遵守內容政策的發布商內容自動具備被考慮的資格。

Google News Publisher Center — answer/15898024（2025-03-20 更新）｜support.google.com/news/publisher-center/answer/15898024｜抓取 2026-08-10

Properly labeled AI content remains fully eligible for monetization.

「正確標示的 AI 內容仍完全具備營利資格」——此句在 Meta 官方文件（transparency.meta.com、facebook.com/business/help、help.instagram.com、meta.com/help/artificial-intelligence）查無逐字或等義的官方保證；它是第三方的改寫或詮釋。Meta 實際條文只說：標示提供透明度，且 AI 內容仍受社群守則與各營利政策約束（原創性門檻見 8.9／8.10）。

查證方法：2026-08-10 於上述四個 Meta 官方網域逐一檢索該句，0 命中。對照官方標示條文：facebook.com/business/help/1010479435004531＋transparency.meta.com/governance/tracking-impact/labeling-ai-content/

Following a recommendation from the Board in February 2024 and ongoing policy development and engagement with external stakeholders, Meta began adding labels to a wider range of video, audio and image content to help users understand when AI is used to create content beginning in May 2024. These labels are now displayed as "AI Info."

可考的官方事實是：2024 年 5 月起 Meta 對更大範圍的影音圖內容加標，標籤現名「AI Info」——這是透明標示制度，不是營利保證。

Meta — Transparency Center: Labeling AI Content｜transparency.meta.com/governance/tracking-impact/labeling-ai-content/｜抓取 2026-08-10

Inauthentic mentions 不在 Google Spam policies（2026-05-15 版）的任何類目清單中。

現行垃圾政策頁共 20 類目（見第 2 節），其中沒有名為「Inauthentic mentions」的類目——把它說成「Google 新增的違規類目」是錯的。

查證方法：2026-08-11 全文抓取 spam policies 頁逐字檢索，0 命中。頁面：developers.google.com/search/docs/essentials/spam-policies

ignore tactics like ... pursuing inauthentic mentions

Google 官方 AI optimization guide 摘要確有此語——勸網站主忽略「追逐不真實提及（pursuing inauthentic mentions）」這類手法。因此「此詞完全是第三方杜撰」的說法（包括本研究第一輪的判定）需要更正：詞是官方的。

Google — AI optimization guide（2026-05-15 加入官方文件；頁面最後更新 2026-07-10）｜developers.google.com/search/docs/fundamentals/ai-optimization-guide｜抓取 2026-08-11

「inauthentic mentions」＝Google 勸避免的 AEO/GEO 手法描述；不是垃圾政策類目、不是獨立違規名稱、也沒有對應的處罰條款。

正確用法：可以說「Google 在 AI 優化指南中勸網站主別追逐不真實提及」；不可以說「Google 新增了 inauthentic mentions 垃圾政策」。另注意：該詞的最初上線日無法從現行頁倒推——已知的是指南於 2026-05-15 加入變更紀錄、現行頁最後更新 2026-07-10。

綜合查證：spam policies（0 命中）＋ AI optimization guide（命中）＋官方 changelog：developers.google.com/search/updates｜查證日 2026-08-11