THE SHORT VERSION
- Meta extended the effective date of its updated Commercial Terms and Self-Serve Ad Terms from 30 October to 20 November after coordinated pushback from publishers, advertisers and trade groups.Meta 把新版商業條款與自助廣告條款的生效日從 10 月 30 日延到 11 月 20 日，此前出版者、廣告主與貿易團體協同反對。
- The terms authorise Meta to use automated software such as web scrapers and crawlers to access business websites in order to provide, develop, improve, research and maintain advertising and business products.條款授權 Meta 使用網路爬蟲等自動化軟體存取企業網站，用於提供、開發、改善、研究與維護廣告及商業產品。
- Publishers asked EU and UK competition authorities to investigate, a group of Norwegian outlets sued, and Meta says implementation could slip again if its consultation runs past 20 November.出版者要求歐盟與英國競爭主管機關調查，挪威媒體提起訴訟；Meta 表示若諮詢超過 11 月 20 日，實施可能再延。
A deadline moved, and possibly moved again期限延後了，可能再延一次
Meta has extended the implementation date of its updated Commercial Terms and Self-Serve Ad Terms to 20 November, pushing back a 30 October deadline after coordinated objections from publishers, digital advertisers and trade groups. A company spokesperson said the extension would allow more time to explain to partners and stakeholders what is changing and what is not.Meta 已把新版商業條款與自助廣告條款的實施日期延到 11 月 20 日，推翻了原本 10 月 30 日的期限；此前出版者、數位廣告主與貿易團體協同提出反對。公司發言人表示，延期是為了有更多時間向合作夥伴與利害關係人說明哪些內容會變、哪些不會變。
The new date may not be final. Meta spokesperson Andy Stone said the company does not plan to make policy changes without a full consultation with industry groups and leaders, and that if the consultation runs beyond 20 November it would delay implementation again. Until then, he said, the existing terms remain in place. The wording of the updated terms has not been withdrawn.新日期未必是最終版本。Meta 發言人 Andy Stone 表示，公司不打算在未與產業團體及領導者充分諮詢前更改政策，如果諮詢延續到 11 月 20 日之後，就會再次延後實施。他說，在那之前現行條款仍然有效。新版條款的文字並未被撤回。
What the terms actually authorise條款到底授權了什麼
The dispute is about a clause, not a press release. The updated terms state that Meta has the authority to use automated software, described as web scrapers, bots, robots, spiders, crawlers and user agents, to access the websites of businesses that use its products in order to provide, develop, improve, research and maintain advertising and business products.爭議的核心是一條條文，而不是新聞稿。新版條款載明，Meta 有權使用自動化軟體，文中列舉網路爬蟲、bots、robots、spiders、crawlers 與 user agents，存取使用其產品之企業的網站，用於提供、開發、改善、研究與維護廣告及商業產品。
Meta says the update streamlines agreements and reduces overlap across its business tools, and that it does not grant unlimited content rights. The company also changed the text earlier this month, removing the word train and adding the reference to advertising and business products, which is the language publishers are now reading closely.Meta 表示，這次更新是為了簡化合約、減少旗下商業工具之間的重疊，並未授予無限的內容權利。公司也在本月稍早修改文字，移除「訓練」一詞，加入關於廣告與商業產品的表述，這正是出版者現在仔細檢視的語言。
Publishers went to regulators, and one senator went to Zuckerberg出版者找上監管機構，參議員寫信給祖克柏
The response was organised. Publishing coalitions including the News/Media Alliance and the European Publishers Council demanded that the terms be withdrawn or suspended, arguing that the language allowed Meta to index external business sites and use the content for AI training and product development without explicit compensation. Industry bodies filed formal complaints with antitrust and competition watchdogs in the European Union and the United Kingdom, and a group of Norway-based outlets sued the company.反彈是有組織的。包括 News/Media Alliance 與歐洲出版者理事會在內的出版聯盟，要求撤回或暫停條款，主張這些文字讓 Meta 得以索引外部企業網站，並在沒有明確補償的情況下把內容用於 AI 訓練與產品開發。產業團體向歐盟與英國的反壟斷及競爭主管機關提出正式申訴，一群挪威媒體也對公司提起訴訟。
California Senator Adam Schiff became the second lawmaker to call on Meta to walk the changes back, in a letter to chief executive Mark Zuckerberg. The breadth of these updated provisions is extraordinary, Schiff wrote, arguing that Meta asserted an alarming opening to bypass paywalls, registration gates and digital subscriber barriers, and that the terms provide no mechanism for businesses to opt out of scraping while keeping their account presence.加州參議員 Adam Schiff 成為第二位要求 Meta 收回變更的民意代表，他致函執行長祖克柏。Schiff 寫道，這些更新條文的範圍「極不尋常」，主張 Meta 藉此開了一道令人警惕的門，得以繞過付費牆、註冊閘門與數位訂閱屏障；他並指出，條款沒有提供任何機制讓企業在不放棄帳號的情況下退出抓取。
Meta says this is not about scraping news sitesMeta 說這不是針對新聞網站
Meta's defence rests on scope and on an existing control. Stone has said the terms are not about scraping news sites and that the company will respect content owners' preferences expressed in robots.txt, the file websites use to tell crawlers what they may fetch. The company points businesses to robots.txt as the standard way to manage crawler access.Meta 的辯護集中在範圍與既有控制手段。Stone 表示，這些條款不是針對抓取新聞網站，公司會尊重內容所有者在 robots.txt 中表達的偏好；robots.txt 是網站告訴爬蟲哪些內容可抓的檔案。公司也把 robots.txt 指為管理爬蟲存取的標準方式。
Critics say that control is weaker than it sounds. Robots.txt preferences are not legally binding, AI companies have regularly ignored them, and the updated terms do not mention robots.txt at all. Without language in the contract codifying a commitment to respect those signals, Schiff wrote, users lack any reliable means to protect their content. That gap, rather than the crawl itself, is what the consultation now has to close.批評者認為這個控制比聽起來薄弱。robots.txt 的偏好不具法律約束力，AI 公司經常忽略它，而新版條款根本沒有提到 robots.txt。Schiff 寫道，如果合約中沒有明文寫下尊重這些訊號的承諾，使用者就沒有任何可靠方式保護自己的內容。這個缺口，而不是抓取本身，正是接下來諮詢要補上的部分。
What it means for a business that advertises on Meta對在 Meta 投放廣告的企業意味著什麼
The terms are not aimed only at news publishers. They govern businesses that use Meta's advertising and business tools, which is why digital advertisers and trade groups joined the objections alongside media companies. For a retailer that buys Instagram ads, the practical question is narrow: does keeping an account and a page mean accepting that Meta may crawl the company's own website for product development, and what happens to the account if it says no.這些條款並非只針對新聞出版者。它們規範的是使用 Meta 廣告與商業工具的企業，這也是數位廣告主與貿易團體會與媒體公司一同反對的原因。對一家購買 Instagram 廣告的零售商而言，實際問題很窄：保留帳號與粉絲頁，是否就等於接受 Meta 可以為了產品開發而抓取公司自家網站；如果拒絕，帳號又會怎樣。
Meta's answer is that businesses can manage crawler access through robots.txt, the standard file that tells crawlers what they may fetch, and that the update does not grant unlimited rights. The critics' answer is that the terms themselves contain no opt-out that keeps an account in place, so a business that wants to refuse the crawl has to rely on a signal the contract does not mention and that AI companies have ignored before. That is the gap the consultation is meant to address, and it is why the 20 November date matters beyond the news industry.Meta 的答案是，企業可以透過 robots.txt 管理爬蟲存取，這是告訴爬蟲哪些內容可抓的標準檔案，而且這次更新並未授予無限權利。批評者的答案是，條款本身沒有提供任何能保住帳號的退出機制，因此想拒絕抓取的企業只能依賴一個合約沒提到的訊號，而 AI 公司過去經常忽略它。這就是諮詢要處理的缺口，也是 11 月 20 日這個日期為何不止關係到新聞業。
What to watch before 20 November11 月 20 日之前要觀察什麼
Regulators are already watching. The European Commission said it welcomed the delay and that it had sent requests for information to companies that have not yet published summaries about the data used to train their AI models. That puts the terms inside a broader EU transparency exercise, not only a contract dispute between Meta and publishers.監管機構已經在關注。歐盟執委會表示歡迎延期，並指出已向尚未公布 AI 模型訓練資料摘要的公司發出資訊請求。這使該條款被納入更廣泛的歐盟透明度作業，而不只是 Meta 與出版者之間的合約爭議。
Three things are worth tracking. Whether the clause text changes at all, whether an opt-out mechanism appears that is more than a pointer to robots.txt, and whether the 20 November date holds. For context, this site has reported on Meta's $97.5 billion AI equity raise, on a privacy case involving Meta's Muse marketplace, and on how AI vendors package enterprise agreements in our report on OpenAI's ServiceNow partnership. Until the contract language changes, the delay buys time, not consent.有三件事值得追蹤：條文文字是否真的修改、是否出現比指向 robots.txt 更實質的退出機制、以及 11 月 20 日這個日期是否守住。相關背景方面，本網站曾報導 Meta 975 億美元 AI 增資、涉及 Meta Muse 市集的隱私案，以及 AI 供應商如何包裝企業協議的 OpenAI 與 ServiceNow 合作。在合約文字改變之前，延期買到的是時間，不是同意。
Sources & reporting limits来源与核查范围
Follow Policy & safety to find related stories on your next visit.关注政策与安全，下次访问时阅读相关文章。
