豆包手机助手正式调整 限制AI刷流量、代打游戏、付款交易

Core Viewpoint - The "Doubao Mobile Assistant" has announced restrictions on its AI capabilities, focusing on three specific scenarios: score brushing, financial applications, and gaming usage [1][3]. Group 1: Product Launch and Features - Doubao Mobile Assistant was launched on December 1, in collaboration with ZTE, using the Nubia M153 model as a customized technology showcase [3]. - The core selling point of the AI assistant is its ability to automate daily app operations, such as ordering food, booking flights, and price comparison shopping [3]. - Within four days of its release, the AI phone was sold out, indicating strong market demand [3]. Group 2: Restrictions and Compliance Issues - Users reported using the Doubao Mobile Assistant for activities like earning rewards on platforms such as Pinduoduo and Douyin, which led to the implementation of restrictions by Doubao [3][4]. - The assistant faced challenges in financial applications, with issues reported in apps like Taobao and major banks, where it could not log in or complete payments [3]. - Doubao stated that while user authorization is required for sensitive operations, it has temporarily disabled capabilities for certain apps as a precaution [3]. Group 3: Security and Privacy Concerns - WeChat users experienced forced logouts when using the Doubao Mobile Assistant, attributed to security risk control measures [4]. - The technology behind the assistant relies on screen reading and simulated operations, which many apps have not authorized, leading to potential security risks [4]. - Experts highlighted that the assistant's screen reading capability bypasses the sandbox mechanism designed to isolate app data, raising privacy concerns [4][5]. Group 4: Industry Implications - The ability of AI to replace user actions could disrupt the real traffic and monetization models of apps, affecting user data security [5]. - Doubao is actively seeking deeper communication with application vendors to establish clearer and more predictable rules, aiming to avoid blanket restrictions on reasonable AI usage by users [5].