华盛顿本周聚集了来自300多家新闻出版商的高管，他们正在游说国会通过一项法案，禁止使用AI“隐形机器人”。新闻/媒体联盟（News/Media Alliance）是新闻和杂志出版商的行业组织，Digiday周二报道说，该联盟组织了此次行程。
Executives from more than 300 news publishers are in Washington this week to lobby Congress for a bill that would ban AI “stealth bots”. The News/Media Alliance, a trade body for news and magazine publishers, organised the trip, Digiday reported on Tuesday.
此次行程的成员包括康德斯·纳斯特（Condé Nast）首席执行官罗杰·林奇（Roger Lynch）、赫斯特杂志（Hearst Magazines）总裁德比·奇里切拉（Debi Chirichella）以及《美国今日报》（USA Today）联合首席执行官迈克·里德（Mike Reed）。他们将会见参议员、众议院其他成员及其工作人员，Digiday的萨拉·瓜格里奥（Sara Guaglione）写道。周三，联盟与美国报纸协会（America’s Newspapers）将为出版商和立法者举办一整天的节目。
Those making the trip include Condé Nast CEO Roger Lynch, Hearst Magazines president Debi Chirichella and USA Today Co CEO Mike Reed. They are meeting senators, other members of Congress and their staff, Digiday’s Sara Guaglione wrote. On Wednesday, the alliance and America’s Newspapers are hosting a day of programming for publishers and lawmakers.
《隐形机器人禁止法案》（Stealth Bot Prohibition Act，H.R. 9915）将隐形机器人定义为在未先披露身份和目的的情况下访问网站的机器人。这包括未通过准确的用户代理字符串识别自身。它还涵盖未说明内容用途的情况，例如搜索索引、AI训练或检索增强生成。
The Stealth Bot Prohibition Act (H.R. 9915) defines a stealth bot as one that accesses a site without first disclosing who it is and why. That includes failing to identify itself through an accurate user-agent string. It also covers failing to state what the content is for, such as search indexing, AI training or retrieval augmented generation.
该法案禁止以可能损害或给网站带来负担的方式使用隐形机器人。它还禁止将机器人伪装成人类用户，以用于生成式人工智能。
The bill bans using stealth bots in ways likely to damage or burden a website. It also bans disguising a bot as a human user for use with generative AI.
联邦贸易委员会（FTC）可对每次违规行为处以最高53,000美元的民事罚款，且每年按通胀调整。州总检察长也可以提起诉讼并为其居民寻求赔偿。该规则将在法案成为法律后180天生效。
The Federal Trade Commission could seek civil penalties of up to $53,000 per violation, adjusted each year for inflation. State attorneys general could also sue and seek damages for their residents. The rules would take effect 180 days after the bill becomes law.
共和党众议员劳雷尔·李（Laurel Lee）和佛罗里达州的古斯·比利拉基斯（Gus Bilirakis），以及北卡罗来纳州的民主党人瓦莱丽·福希（Valerie Foushee）于七月提出了该法案。它目前在众议院能源与商务委员会审议。
Republican Reps Laurel Lee and Gus Bilirakis of Florida and Democrat Valerie Foushee of North Carolina introduced the bill in July. It sits with the House Energy and Commerce Committee.
“今天，网站运营商往往无法得知谁在访问他们的系统，以及是否有自动化工具在以虚假身份收集他们的内容，”李在当时表示。
“Today, website operators are too often left in the dark about who is accessing their systems and whether automated tools are collecting their content under false pretenses,” Lee said at the time.
新闻/媒体联盟称该法案旨在为全国统一制定相关标准。密苏里州、内布拉斯加州、田纳西州、德克萨斯州和犹他州也已提出了类似的法规；纽约州则于今年6月通过了自己的《隐形爬虫禁令》。该联邦法案明确表示，该法案不会影响其他联邦或州法律所规定的权利或救济措施。
The News/Media Alliance calls the bill an attempt to set one standard for the whole country. Missouri, Nebraska, Tennessee, Texas and Utah have proposed similar rules. New York passed its own Stealth Crawler Prohibition Act in June. The federal bill’s text says it does not affect rights or remedies under other federal or state laws.
出版商为何支持这项法案？根据 Digiday 援引的 Cloudflare 数据，目前机器人产生的网络流量已占互联网总流量的 60% 以上。Cloudflare 的首席执行官 Matthew Prince 告诉 The Verge，机器人的访问量可能是人类访问量的 1000 倍。
Why publishers want it Bots now make up more than 60% of internet traffic, according to Cloudflare data cited by Digiday. Cloudflare CEO Matthew Prince told The Verge that bots could reach 1,000 times human traffic.
TollBit 在 2026 年上半年统计发现，其分析的网站中存在超过 220 亿次由人工智能程序发起的爬取行为；在其监测的出版商网站中，人工智能程序的访问量与人类访问量的比例在 2025 年内增长了六倍以上，到第四季度时这一比例已降至 1:31。
TollBit counted more than 22 billion AI bot scrapes across the sites it analysed in the first half of 2026. On the publisher sites it tracks, the ratio of AI bot visits to human visits rose more than sixfold during 2025. By the fourth quarter, it stood at one in 31.
该联盟的首席执行官 Danielle Coffey 表示，许多参与该法案制定的高管都与相关立法者所代表的社区有着密切联系。她指出：“许多出版商都面临着同样的问题：大量机器人访问他们的网站，导致他们的网站不堪重负；即使网站规模较大，也无法有效应对这些流量；更糟糕的是，这些机器人还会窃取他们的内容，从而破坏整个市场秩序。”Lynch 早在 7 月份就表示支持这项法案。
Danielle Coffey, the alliance’s CEO, said many of the executives have deep ties to the communities lawmakers represent. > “They’re all having the same experiences. Too many of these bots are coming on their sites. They can’t handle the traffic because they’re too small. If they can even handle it, if they’re larger, it’s stealing their content. It’s undermining the marketplace.”Lynch backed the bill in July.
Coffey 进一步解释说：“人工智能公司利用伪装后的机器人程序来窃取原创新闻内容，且完全无需承担任何责任；它们利用我们的内容来训练和运行自己的系统，进而直接与我们的业务形成竞争。”据 Business Insider 报道，Lynch 周三向 Condé Nast 的员工宣布他将离职，转任 Mattel 的首席执行官。
“AI companies are using disguised bots to scrape and steal original journalism with zero accountability. They use our work to train and operate their systems that then directly compete with us.”On Wednesday, Lynch told Condé Nast staff he is leaving to become CEO of Mattel, Business Insider reported.
Coffey 希望这项法案能成为市场所需的“基础性整治措施”：如果隐形爬虫程序必须公开自己的身份，出版商就可以将其屏蔽，从而与人工智能公司签订许可协议。她表示：“这项法案只是实现目标的手段而已。”
Coffey said she hopes the bill can be the “foundational cleanup” the market needs. If stealth crawlers must identify themselves, publishers can block them, she said. Then they can sign licensing deals with AI companies.
David Buttle 是一个致力于推动出版业人工智能发展的联盟（SPUR）的创始人，他认为该法案将提高隐形爬虫程序的运营成本和法律风险。
“The bill is a means to an end,” Coffey told Digiday. David Buttle founded SPUR, a publisher AI coalition. He said the bill could raise the cost and legal risk of stealth scraping.
“公司目前正在为内容付费，但他们是付费给中间抓取服务来完成，而不是付给出版商。”Reed 在给 Digiday 的采访中表示，这项法案是“在获取和使用出版商内容方面迈向更大透明度的有意义一步”。
“Companies are currently paying for content but they’re paying intermediary scraping services to do this, rather than publishers.”Reed told Digiday the bill is “a meaningful step toward greater transparency” on how publisher content is accessed and used.
“通过团结一致、以统一的声音发声，我们有更大的机会为有助于创建更健康信息生态系统的立法争取支持，”Chirichella 说。
“By coming together and speaking with one voice, we have a stronger opportunity to build support for legislation that will help create a healthier information ecosystem,” Chirichella said.
Coffey 说，今年的飞行会议的高管人数大约是去年的两倍，并且多了一天的节目安排。去年会议的讨论激发了这项法案。
This year’s fly-in has roughly twice as many executives as last year’s and an extra day of programming, Coffey said. Conversations from last year’s meetings inspired the bill.
其他平台也在采取抓取行动。Reddit 本周表示，将于 11 月 13 日关闭 RSS 订阅，理由是 AI 抓取。7 月，Cloudflare 给 AI 爬虫设定了 9 月的最后期限，要求它们向出版商付款，否则将被封锁。
Other platforms are acting on scraping too. Reddit said this week it will shut down RSS feeds on 13 November, citing AI scraping. In July, Cloudflare gave AI crawlers a September deadline to pay publishers or face a block.