特朗普政府允许Anthropic向美国特定机构发布Mythos。
内容来源：https://www.wired.com/story/anthropic-restores-access-to-mythos/
内容总结：
美国政府放宽对Anthropic最强AI模型的限制，允许百余家美国机构重新使用
美国政府近日放宽了对人工智能公司Anthropic最强AI模型“Claude Mythos 5”的出口管制，允许该公司向超过100家美国组织开放访问权限，其中包括大型企业及政府机构。
据《连线》杂志获得的美国商务部长霍华德·卢特尼克致Anthropic联合创始人兼首席计算官汤姆·布朗的信件显示，卢特尼克告知这家AI实验室，因其“已确定相关安全保障措施到位”，将允许部分受信任的合作伙伴使用Mythos模型。这一消息由Semafor率先报道。
卢特尼克在信中写道：“Anthropic已与美国政府合作，共同应对与受管制模型相关的风险。这些努力取得了显著进展。”
然而，美国政府并未批准该模型更广泛的推广，也未就面向消费者的Claude Fable 5模型作出任何表态。Anthropic在发布Fable 5时配备了额外的安全保护措施。卢特尼克在信中强调，其在6月12日发出的初始指令中列出的其他要求依然有效。
Anthropic发言人爱德华多·马亚·席尔瓦在给《连线》的声明中表示：“我们已收到美国政府通知，我们最强的网络安全模型Mythos 5可以重新部署给一小部分网络防御者和基础设施提供商。我们正尽快为获批的提供商开通权限，恢复他们对Mythos 5的使用。我们对这一进展感到高兴，并将继续与政府合作，扩大Mythos 5的访问范围，并争取让Fable 5重新向公众开放。”
据知情人士透露，Anthropic仍在与白宫就恢复Fable 5的使用权限进行磋商，相关讨论预计将持续至本周末。该人士表示，双方都希望此次事件的解决能为未来模型发布制定一个持久性的政策框架提供参考。
此次部分解禁发生在白宫向Anthropic发出出口管制指令约两周后。当时白宫要求该公司限制外国公民访问Mythos和Fable 5模型，包括在美国工作和生活的人员。作为回应，Anthropic完全关闭了这些模型的访问权限。在最新的信件中，卢特尼克表示，获准使用Mythos的组织现在可以允许其外籍员工访问该模型，Anthropic自身也可以允许其外籍员工使用。
据《连线》此前报道，特朗普政府之所以对Anthropic发布Mythos感到担忧，是因为发现该公司曾授权给一家据信与中国有关联的韩国电信公司。此外，亚马逊和美国国家安全局也分别向白宫提出担忧，认为Fable 5可能被“越狱”利用。这些事件叠加，促使官员们认为必须采取行动。
最近几周，Anthropic派出了网络安全和AI安全团队的高级成员前往华盛顿，与特朗普政府官员会面。除布朗外，Anthropic公共政策主管萨拉·赫克也一直主导着公司与美国商务部的对话。
让Mythos 5重新上线对Anthropic和白宫来说是向前迈出的积极一步，但这一事件引发了关于美国AI政策整体方向的更广泛质疑，尤其是特朗普政府将在多大程度上试图控制未来模型的发布。本周五，OpenAI宣布应特朗普政府的要求，推迟发布其即将推出的GPT 5.6模型。
OpenAI战略未来团队负责人、前白宫AI顾问迪恩·鲍尔在6月16日的一篇博文中表示，Anthropic与特朗普政府的此次摩擦表明，前沿AI模型开发商现在“需要获得政府的明确绿灯”。
Anthropic与白宫的角力已经对这家年轻公司的业务造成了损失。今年早些，该公司曾因在军事承包商如何使用其AI模型方面划定红线而收到供应链风险认定，并就此起诉了特朗普政府。据《连线》此前报道，在白宫于6月12日周五晚发出出口指令后，部分Anthropic投资者在周末加班，以评估此事对公司未来发展的影响。
中文翻译：
| 美国政府已放宽对Anthropic公司最先进AI模型Claude Mythos 5的限制，允许该公司向包括大型企业和政府机构在内的100多家美国组织开放访问权限。 《连线》杂志获得的一封致Anthropic联合创始人兼首席计算官汤姆·布朗的信件显示，美国商务部长霍华德·卢特尼克告知这家AI实验室，由于其“已确认存在适当的安全保障措施”，将允许特定可信合作伙伴访问Mythos模型。Semafor率先报道了这封信件的存在。 卢特尼克在信中写道：“Anthropic已与美国政府合作应对受监管模型的相关风险。这些努力已取得显著进展。” 然而，美国政府并未批准该模型更广泛的推广，也未提及面向消费者的Claude Fable 5模型的命运——这是Anthropic在增加额外安全防护措施后发布的Mythos消费者版本。卢特尼克在信中强调，其6月12日最初指令中列出的其他要求仍然有效。 Anthropic发言人爱德华多·马亚·席尔瓦在给《连线》的声明中表示：“我们已收到美国政府通知，我们最强的网络安全模型Mythos 5可重新部署给一小批网络防御者和基础设施提供商。我们正尽快为获批的供应商开通权限并恢复其Mythos 5访问权。我们很高兴看到这一进展，并将继续与政府合作扩大Mythos 5的访问权限，并推动Fable 5重新面向大众开放。” 据知情人士透露，Anthropic仍在与白宫就恢复Fable 5访问权进行谈判，预计相关讨论将持续至周末。该人士表示，双方均希望此次事件的解决方案能为未来模型发布提供长期政策框架参考。 |
有线索？ |
|---|---|
| 您是现任或前任Anthropic或美国政府雇员，想聊聊正在发生的事吗？我们期待您的消息。请使用非工作电话或电脑，通过Signal以安全方式联系记者mzeff.88。 |
此次部分恢复访问权距离白宫向Anthropic发出出口管制指令约两周。该指令要求公司限制外国公民（包括在美国工作和生活的人员）访问Mythos和Fable 5模型。作为回应，Anthropic完全禁用了这两个模型的访问权限。在最新信件中，卢特尼克写道，获批使用Mythos的组织现在可允许其外籍员工访问该模型，Anthropic也可对其自身外籍员工采取相同措施。
据《连线》此前报道，特朗普政府在获悉Anthropic向一家其认为与中国存在关联的韩国电信公司提供访问权限后，开始对Mythos的发布感到担忧。亚马逊和国家安全局也分别向白宫提出Fable 5可能被破解的风险，这些事件的叠加使官员们确信必须采取行动。
近几周，Anthropic派出网络安全和AI安全团队的高级成员前往华盛顿，与特朗普政府官员会面。除布朗外，Anthropic公共政策负责人莎拉·赫克也主导了公司与美国商务部的谈判。
Mythos 5恢复上线对Anthropic和白宫而言是积极一步，但这一系列事件引发了关于美国AI政策总体方向的更广泛质疑，尤其是特朗普政府将在多大程度上试图控制未来模型的发布。周五，OpenAI宣布应特朗普政府要求，将推迟发布即将推出的GPT 5.6模型。
OpenAI战略未来团队负责人、前白宫AI顾问迪恩·鲍尔在6月16日的博客文章中表示，Anthropic近期与特朗普政府的摩擦表明，前沿AI模型开发商如今“需要获得政府的明确许可”。
Anthropic与白宫的博弈已对这家年轻公司的业务造成损失。今年早些时候，该公司在试图界定军事承包商如何使用其AI模型时，因获得的供应链风险认定起诉了特朗普政府。据《连线》此前报道，在白宫于6月12日周五晚间发出出口指令后，部分Anthropic投资者整个周末都在评估这对公司未来的影响。
英文来源：
| The US government has eased the restrictions it imposed on Anthropic’s most advanced AI model, Claude Mythos 5, allowing the company to grant access to more than 100 US organizations, including large corporations and government agencies. In a letter sent to Anthropic’s cofounder and chief compute officer Tom Brown obtained by WIRED, US Commerce Secretary Howard Lutnick told the AI lab it would permit certain trusted partners to access Mythos because he had “determined that appropriate safeguards are in place.” Semafor first reported the existence of the letter. “Anthropic has worked with the U.S. government to address risks associated with the Covered Models. These efforts have yielded significant progress,” Lutnick wrote. However, the government stopped short of permitting a broader rollout of the model, and said nothing about the fate of Claude Fable 5, the consumer-facing version of Mythos that Anthropic released with significant additional safeguards. Lutnick noted in his letter that the other requirements outlined in the initial directive he sent on June 12 remain in effect. “We received notice from the US government that Mythos 5, our strongest cybersecurity model, can be redeployed to a small group of cyber defenders and infrastructure providers,” Anthropic spokesperson Eduardo Maia Silva said in a statement to WIRED. “We are working to provision the approved set of providers and restore their access to Mythos 5 as quickly as possible. We are pleased to see this progress and continue to work with the government to expand access to Mythos 5 and make Fable 5 available for general use again.” Anthropic is still in discussions with the White House about restoring access to Fable 5, and they are expected to continue over the weekend, according to a person familiar with the matter. Both parties are hopeful the resolution of this incident will help inform a lasting policy framework for future model releases, the person said. |
Got a Tip? |
|---|---|
| Are you a current or former Anthropic or US government employee who wants to talk about what's happening? We'd like to hear from you. Using a nonwork phone or computer, contact the reporter securely on Signal at mzeff.88 |
The partial reinstatement comes roughly two weeks after the White House sent an export control directive to Anthropic that required the company to limit foreign nationals from accessing Mythos and Fable 5, including people working and living in the United States. In response, Anthropic disabled access to the models entirely. In his latest letter, Lutnick wrote that organizations approved to use Mythos may now allow their foreign national employees to access the model, and Anthropic may do the same for its own foreign national employees.
The Trump administration grew concerned about Anthropic’s rollout of Mythos after it learned the company granted access to a South Korean telecommunications firm it believed had ties to China, WIRED previously reported. Amazon and the National Security Agency also separately raised concerns to the White House that Fable 5 could be jailbroken, and the confluence of events convinced officials they needed to take action.
In recent weeks, Anthropic sent senior members from its cybersecurity and AI safety teams to Washington, DC to meet with Trump administration officials. Along with Brown, Anthropic's public policy chief Sarah Heck have been leading the company’s discussions with the US Department of Commerce.
Getting Mythos 5 back online marks a promising step forward for Anthropic and the White House, but the saga has raised broader questions about the overall direction of US AI policy, particularly the extent to which the Trump administration will seek to control future model releases. On Friday, OpenAI announced it was delaying the release of its upcoming GPT 5.6 models in response to a request from the Trump administration.
Dean Ball, head of the strategic futures team at OpenAI and a former White House AI adviser, said in a June 16 blog post that Anthropic’s latest spat with the Trump administration has shown frontier AI model developers they “need an explicit green light from the government now.”
Anthropic’s battles with the White House have been costly for the young company’s business. It sued the Trump administration earlier this year over a supply chain risk designation it received after trying to draw red lines around how military contractors could use its AI models. After the White House sent its export directive on the Friday evening of June 12, WIRED previously reported that some Anthropic investors worked through the weekend to determine what it meant for Anthropic’s corporate future.
文章标题：特朗普政府允许Anthropic向美国特定机构发布Mythos。
文章链接：https://news.qimuai.cn/?post=4459
本站文章均为原创，未经授权请勿用于任何商业用途