苹果 App Store 新规落地 AI 深度伪造迎来最严审核Apple's new App Store rules for in-depth AI counterfeiting usher in the most rigorous review


AI free token/AI免费积分
苹果 App Store 新规落地 AI 深度伪造迎来最严审核
感谢IT之家网友 对的时间点、乌蝇哥的左手 的线索投递!

IT之家 10 月 17 日消息,苹果公司昨日(10 月 16 日)更新《App 审核指南》,在最新修订的 1.2 条款中,加强治理针对 AI 内容创作类应用,将 AI 生成的、未经当事人同意的私密图像明确纳入禁止范围。

IT之家附上修订内容如下:

1.2 使用者生成内容

对于包含使用者生成内容的 App,有特定的难题需要应对,例如知识产权侵权、未经同意传播私密图像、匿名欺凌等。为了避免滥用,包含使用者生成内容或社交网络服务的 App 必须满足以下条件:

科技媒体 9to5Mac 指出,此次调整的直接背景,是近一年围绕 App Store 的多起争议。科技透明度项目曾发现数十款 AI“裸体化”应用,并指出 App Store 的搜索建议和广告也会把使用者导向这类应用;旧金山官员随后要求苹果下架 8 款可生成未经同意私密图像的应用。

另外使用者曾投诉 xAI 的 Grok 聊天机器人被用于制作裸露或性化深度伪造图像;今年早些时候,苹果曾警告 xAI,若继续放任此类内容,将把 Grok 从 App Store 下架。xAI 随后增设防护机制,更新版本才取得批准。

苹果还新增 2.5.19 条准则,禁止应用在审核时隐瞒真实能力、借助后再加入“裸露化”(nudify)等有害能力。这类“先过审、后解锁”的做法被称为“诱饵式营销”,今后可能导致应用下架,开发者也可能被终止 Apple Developer Program 资格。

另外,苹果将禁止助长人口贩卖或剥削儿童的应用的范围,扩展至替代应用分发渠道,并表示对违规应用和开发者“不会犹豫”采取下架或终止合作等措施。

本文来源:https://www.ithome.com/1/014/209.htm
若有版权问题,请来信告知:13951431913#139.com 或在右方提交评论。

IT House October 17th news, Apple yesterday (October 16th) updated the "App Review Guidelines", in the latest revision of Clause 1.2, strengthen governance for AI content creation applications, and explicitly include the prohibition of private images generated by AI without the consent of the parties.

IT Home is accompanied by the following amendments:

1.2 User Generated Content

For apps that contain user-generated content, there are specific challenges that need to be addressed, such as intellectual property infringement, the dissemination of private images without consent, anonymous bullying, and more. To avoid abuse, apps that include user-generated content or social networking services must:

Technology media 9to5Mac pointed out that the immediate background of the adjustment is the many controversies surrounding the App Store in the past year. The Tech Transparency Project has found dozens of AI “nudity” apps, noting that search suggestions and ads on the App Store will also direct users to such apps; San Francisco officials then asked Apple to remove eight apps that generate private images without consent.

Apple's new App Store rules for in-depth AI counterfeiting usher in the most rigorous review

In addition, users have complained that xAI's Grok chatbot is used to create nudity or sexualized deepfakes; earlier this year, Apple warned xAI that it would remove Grok from the App Store if it continued to allow such content. xAI subsequently added protection mechanisms to obtain approval for the updated version.

Apple has also added 2.5.19 guidelines to prohibit apps from hiding real abilities during review and adding harmful abilities such as "nudify" after use. This type of "review first, unlock later" practice is called "decoy marketing" and may lead to app deactivation in the future, and developers may be terminated from the Apple Developer Program.

In addition, Apple expanded its ban on apps that promote human trafficking or the exploitation of children to alternative app distribution channels, and said it would "not hesitate" to take measures such as suspending or terminating cooperation with violating apps and developers.

This article is from: https://www.ithome.com/1/014/209.htm For copyright issues, please write to: 13951431913#139.com or submit a review on the right.

暂无评论,快来发表第一条评论吧!No comments yet. Be the first!

📮 需求咨询📮 Inquiry