广告
加载中

Personal Agent时代来了 千问要成为最懂你的AI

戴菁 2026-09-24 09:08
戴菁 2026/09/24 09:08

邦小白快读

EN
全文速览

Personal Agent时代的核心干货是:AI不再只是问答工具,而是能长期记住你的专属助手。它会在后台持续存在,基于你过去提到的信息主动为你办事,比如记得食谱和朋友过敏史,自动帮你准备聚会。

重点信息包括:千问和Muse是两大代表产品,Muse上线12天下载量280万,日活渗透率23%;千问上线10个月复杂任务需求增长超5倍,62%的简短提问需要中长期上下文。

实操干货是:你不需要学习提示词或复杂工作流,只需像聊天一样说出需求,AI会结合你的历史数据、设备信息和授权账户来给出个性化建议。比如告诉千问理财目标,它会结合行情和你的持仓数据给出分析;健康场景它能记得你孩子的年龄和体检异常项。

对普通人的价值在于:AI正在从服务具体任务转向服务复杂的人,未来每个人都能拥有一个真正属于自己的个人助理,安排日常生活、做选择、处理长期目标,而且这种服务正在普惠化。

品牌营销新机会:Personal Agent能长期记忆用户偏好和场景关联,品牌可通过AI在恰当时机推荐商品或服务。例如Muse会结合用户保存的食谱和朋友过敏信息自动生成聚会方案,这为品牌提供了精准触达消费者的新场景。

用户行为观察:2025年凯捷调查显示36%消费者希望获得自动化个性化GenAI支持,61%希望AI自动补货常购商品;埃森哲调查称80%的人愿将至少一项日常任务委托AI代办。这意味着品牌需适应AI代办的购物决策模式。

消费趋势与产品研发:用户需求从简单问答转向复杂任务和长期陪伴,品牌应开发适合AI理解与推荐的产品形态,重视场景关联数据,如健康与运动、子女教育与理财的交叉需求。

渠道建设:千问开放平台已接入租房、求职、运动、物流、游戏等20多个领域智能体,品牌可考虑接入此类个人智能生态,借助阿里电商、支付、地图等接口实现从理解到行动的闭环。

政策解读与趋势:Personal Agent成为AI行业高光方向,Meta的Muse上线12天下载量280万、带动股价暴涨11%,阿里千问加速打造Personal Agent,这表明专属智能助手是新的增长市场。卖家应关注AI推荐驱动的消费场景。

消费需求变化:用户希望AI记住偏好并主动代办,61%的人希望自动补货常购商品,80%的人愿委托日常任务。这意味着复购型、计划型商品的销售可通过AI代理实现自动化。

事件应对与机会:千问已具备AI购物、打车、订机票酒店等能力,卖家可借助千问开放平台接入外部服务,获得来自阿里生态的流量和服务接口。开放平台智能体已扩展到租房、求职、运动等20多领域,合作空间大。

风险提示:个人AI对数据隐私和保障措施要求高,卖家需确保商品和服务能兼容AI的个性化推荐逻辑,避免因不透明或不合规被AI筛选排除。

可学习点:关注Personal Agent的长期记忆和跨场景能力,卖家可在产品描述中突出结构化信息,便于AI理解与调用,从而提升被推荐概率。

产品生产和设计需求:Personal Agent时代要求智能硬件设备接入个人状态模型,例如智能手表、动态血糖仪等健康设备。工厂可关注多终端设备的数据采集与连接需求,开发能配合AI助手的硬件产品。

商业机会:千问开放平台已覆盖租房、求职、运动、物流、游戏等20多个领域,工厂可通过智能硬件或供应链服务切入这些场景,提供AI可调用的实体服务。

数字化与电商启示:AI助手能连接手机、手表、PC、AI眼镜等终端,工厂应推动产品数字化、联网化,使硬件产生的数据能够被Personal Agent理解与调用,从而进入个人智能生态。

生产模式变化:个人需求不等于简单需求,跨场景关联(如运动与健康、旅行与消费)要求产品具备场景适配性。工厂在设计产品时需考虑多场景数据融合,避免孤立地生产单一功能设备。

行业发展趋势:Personal Agent封装了陪伴+办事两大能力,成为AI应用新方向。Meta的Muse和阿里千问是代表性案例,行业正从单一任务服务走向对复杂人的连续理解。

新技术与客户痛点:客户痛点是个人Context分散且互不相通。千问正在从三个维度解决:通用Context(你是谁)、领域级Context(你正在做什么)、多终端Context(你此刻经历什么)。服务商可围绕这些维度提供数据整合、场景理解技术。

解决方案:建立不断更新的个人状态模型是核心。服务商可开发跨时间、跨场景的Context组织技术,将用户在不同App、设备、服务中的信息连接起来,形成连续理解。

客户需求数据:36%消费者希望获得自动化个性化GenAI支持,61%希望AI自动补货,80%愿委托AI代办。这为服务商提供了明确的市场方向,可开发面向个人长期生活管理的Agent服务。

代表企业与接口:千问依托Qwen3.8系列大模型和阿里生态(电商、支付、地图、出行),服务商可通过千问开放平台接入外部服务,获得实际落地接口。

商业对平台的需求:Personal Agent需要跨App、跨服务、跨终端调用数据和执行任务,平台需提供统一的数据接入和行动接口。阿里通过电商、支付、地图、出行等生态服务,以及千问开放平台,正是满足这一需求的举措。

平台最新做法:千问开放平台智能体已扩展到20多个领域(租房、求职、运动、物流、游戏等),并接入智能手表、动态血糖仪等设备。平台商可参考这种开放生态策略,吸引更多服务接入,增强用户粘性。

招商与运营管理:平台需支持不同领域智能体的互操作,允许调用用户授权的个人数据。千问在健康、理财等领域已打通智能体和用户授权数据,平台商应设计清晰的数据授权机制和领域接入标准。

风向规避:个人数据隐私和保障措施是关键,埃森哲调查显示80%的人愿意在保障措施到位后委托AI代办,平台需建立透明、合规的隐私框架,避免因数据滥用引发信任危机。

运营数据参考:千问上线10个月复杂任务需求增长超5倍,沟通轮次5-8轮,62%简短提问需中长期上下文。平台运营需重视长期记忆能力和用户档案画像的建设。

产业新动向:Personal Agent成为AI应用新趋势,标志着AI从服务具体任务转向服务复杂的人。Meta的Muse和阿里千问是标志性案例,Muse的下载和渗透率数据、千问的用户需求结构变化提供了实证。

新问题:个人Context具有跨时间、跨场景、跨终端的连续变化特征,如何将散落的身份、状态、关系持续拼合成完整的人,是Personal Agent的核心难题。现有AI仅在单一场景处理Context,无法满足跨场景关联需求。

政策与法规建议:个人数据授权与隐私保障是AI代办的信任前提,80%的人愿意在保障措施到位后委托AI代办。研究者需关注如何设计合规的数据访问、授权和遗忘机制,平衡个性化服务与数据安全。

商业模式:Personal Agent可基于长期用户档案提供订阅式或按需付费的个性化服务,千问已通过开放平台+生态服务体系构建了从理解到行动的闭环,这种平台化模式可能成为重要商业方向。

理论基础:文章提出个人状态模型概念,从通用Context、领域级Context、多终端Context三个维度构建连续理解。研究者可深入探讨此模型在不同生活场景中的动态更新机制及评估方法。

返回默认

声明:快读内容全程由AI生成,请注意甄别信息。如您发现问题,请发送邮件至 run@ebrun.com 。

我是 品牌商 卖家 工厂 服务商 平台商 研究者 帮我再读一遍。

Quick Summary

The core value proposition of the Personal Agent era is that AI is no longer just a Q&A tool, but a dedicated assistant that remembers you over the long term. It operates persistently in the background and proactively handles tasks based on information you have mentioned before—such as remembering a recipe and a friend's allergy history, and automatically preparing a gathering for you.

Key facts include: Qwen and Muse are the two representative products. Muse reached 2.8 million downloads in 12 days with a 23% daily active user penetration rate; Qwen saw a more than 5x increase in complex task demand in 10 months, and 62% of short queries require mid-to-long-term context.

Practical guidance: You don't need to learn prompt engineering or complex workflows. Simply state your needs as if chatting, and the AI will combine your historical data, device information, and authorized accounts to offer personalized recommendations. For example, tell Qwen your financial goal, and it will provide analysis based on market conditions and your portfolio data; in a health scenario, it can remember your child's age and abnormal checkup results.

The value for ordinary people is that AI is shifting from serving specific tasks to serving complex individuals. In the future, everyone can have a truly personal assistant to arrange daily life, make choices, and handle long-term goals—and this service is becoming more accessible to all.

New opportunities for brand marketing: Personal Agents can remember user preferences and contextual associations over the long term, allowing brands to recommend products or services through AI at the right moment. For example, Muse can combine a user's saved recipes and friends' allergy information to automatically generate a gathering plan, providing brands with a new scenario for precision-reaching consumers.

User behavior observations: A 2025 Capgemini survey shows that 36% of consumers want automated, personalized GenAI support, and 61% want AI to automatically replenish frequently purchased items; Accenture research indicates that 80% of people are willing to delegate at least one daily task to AI. This means brands need to adapt to AI-mediated shopping decision patterns.

Consumption trends and product development: User needs are shifting from simple Q&A to complex tasks and long-term companionship. Brands should develop product formats that are easy for AI to understand and recommend, and pay attention to context-related data, such as cross-domain needs across health and sports, children's education and financial management.

Channel building: Qwen's open platform has already integrated agents across more than 20 domains, including rental housing, job seeking, sports, logistics, and gaming. Brands can consider joining this personal intelligence ecosystem and leveraging Alibaba's e-commerce, payment, and mapping interfaces to close the loop from understanding to action.

Policy interpretation and trends: Personal Agent has become a high-profile direction in the AI industry. Meta's Muse reached 2.8 million downloads in 12 days and drove its stock price up 11%; Alibaba's Qwen is accelerating the development of its Personal Agent. This suggests that dedicated intelligent assistants represent a new growth market. Sellers should pay close attention to AI-recommendation-driven consumption scenarios.

Changes in consumer demand: Users want AI to remember preferences and proactively handle tasks. 61% want automatic replenishment of frequently purchased items, and 80% are willing to delegate everyday tasks. This means sales of repurchase-based and planned goods can be automated through AI agents.

Opportunities and responses: Qwen already has capabilities such as AI shopping, ride-hailing, flight and hotel booking. Sellers can leverage Qwen's open platform to connect external services and gain traffic and service interfaces from Alibaba's ecosystem. The open platform's agents have expanded to more than 20 domains including rental housing, job hunting, and sports, leaving substantial room for collaboration.

Risk warning: Personal AI places high demands on data privacy and safeguards. Sellers need to ensure that their products and services are compatible with AI's personalized recommendation logic, and avoid being excluded by AI filtering due to opacity or non-compliance.

Key takeaways: Focus on Personal Agent's long-term memory and cross-scenario capabilities. Sellers can highlight structured information in product descriptions to make it easier for AI to understand and invoke, thereby increasing the probability of being recommended.

Product production and design requirements: The Personal Agent era requires smart hardware devices to connect to personal state models, such as smart watches, continuous glucose monitors, and other health devices. Factories should pay attention to the data collection and connectivity needs of multi-device terminals and develop hardware products that can work with AI assistants.

Business opportunities: Qwen's open platform already covers more than 20 domains, including rental housing, job seeking, sports, logistics, and gaming. Factories can enter these scenarios through smart hardware or supply chain services, providing tangible services that AI can invoke.

Digitalization and e-commerce insights: AI assistants can connect phones, watches, PCs, AI glasses, and other terminals. Factories should push for product digitalization and networking so that the data generated by hardware can be understood and invoked by Personal Agents, thereby entering the personal intelligence ecosystem.

Changes in production models: Personal needs are not equal to simple needs. Cross-scenario associations (such as sports and health, travel and consumption) require products to be scenario-adaptive. When designing products, factories need to consider multi-scenario data integration and avoid producing isolated single-function devices.

Industry development trends: Personal Agent encapsulates two core capabilities—companionship and task execution—and has become a new direction for AI applications. Meta's Muse and Alibaba's Qwen are representative cases. The industry is moving from single-task services to continuous understanding of complex individuals.

New technologies and customer pain points: The customer pain point is that personal context is fragmented and not interconnected. Qwen is addressing this from three dimensions: general context (who you are), domain-level context (what you are doing), and multi-device context (what you are experiencing right now). Service providers can offer data integration and scenario understanding technologies around these dimensions.

Solutions: Building a continuously updated personal state model is the core. Service providers can develop context organization technologies that span time and scenarios, connecting information from users' different apps, devices, and services to form continuous understanding.

Customer demand data: 36% of consumers want automated personalized GenAI support, 61% want AI to automatically replenish goods, and 80% are willing to delegate tasks to AI. This provides a clear market direction for service providers to develop agent services for individuals' long-term life management.

Representative companies and interfaces: Qwen relies on the Qwen 3.8 series large models and Alibaba's ecosystem (e-commerce, payment, maps, mobility). Service providers can access external services through Qwen's open platform and obtain practical deployment interfaces.

Business requirements for platforms: Personal Agents need to invoke data and execute tasks across apps, services, and devices. Platforms must provide unified data access and action interfaces. Alibaba's move to offer e-commerce, payment, mapping, and mobility ecosystem services, along with the Qwen open platform, is exactly a response to this need.

Latest platform practices: Qwen's open platform agents have expanded to more than 20 domains (rental housing, job seeking, sports, logistics, gaming, etc.) and have integrated devices such as smart watches and continuous glucose monitors. Platform vendors can reference this open ecosystem strategy to attract more services and enhance user stickiness.

Partner recruitment and operations management: Platforms need to support interoperability of agents across different domains and allow invocation of user-authorized personal data. Qwen has already connected agents with user-authorized data in areas such as health and finance. Platform vendors should design clear data authorization mechanisms and domain access standards.

Risk avoidance: Personal data privacy and safeguards are critical. Accenture's survey shows that 80% of people are willing to delegate tasks to AI after safeguards are in place. Platforms need to establish transparent, compliant privacy frameworks to avoid trust crises caused by data misuse.

Operational data reference: In 10 months since launch, Qwen has seen more than a 5x increase in complex task demand, 5-8 rounds of interaction, and 62% of short queries requiring mid-to-long-term context. Platform operations should emphasize long-term memory capability and the building of user profiles.

New industry trends: Personal Agent has become a new direction in AI applications, marking a shift from serving specific tasks to serving complex individuals. Meta's Muse and Alibaba's Qwen are landmark cases; Muse's download and penetration data and Qwen's changing user demand structure provide empirical evidence.

New problems: Personal context has the characteristics of continuous change across time, scenarios, and devices. How to continuously integrate scattered identity, status, and relationships into a complete person is the core challenge of Personal Agent. Existing AI only processes context in single scenarios and cannot meet cross-scenario association needs.

Policy and regulatory recommendations: Personal data authorization and privacy protection are the prerequisites for trust in AI delegation. 80% of people are willing to delegate tasks to AI after safeguards are in place. Researchers need to focus on designing compliant data access, authorization, and forgetting mechanisms that balance personalized services with data security.

Business models: Personal Agent can provide subscription-based or on-demand personalized services based on long-term user profiles. Qwen has already built a closed loop from understanding to action through its open platform and ecosystem services. This platform-based model may become an important business direction.

Theoretical basis: The article proposes the concept of a personal state model, which builds continuous understanding from three dimensions: general context, domain-level context, and multi-device context. Researchers can further explore this model's dynamic update mechanisms and evaluation methods in different life scenarios.

Disclaimer: The "Quick Summary" content is entirely generated by AI. Please exercise discretion when interpreting the information. For issues or corrections, please email run@ebrun.com .

I am a Brand Seller Factory Service Provider Marketplace Seller Researcher Read it again.

今年9月,AI行业的高光是Personal Agent给的。

Meta旗下面向普通用户的AI智能体Muse上线12天,下载量280万,不仅全面超越ChatGPT同期数据,还快速登上了美国App Store榜首。

Muse的强劲表现,直接带动Meta的股价本周一暴涨11%,并为美国科技行业注入新的士气。

时隔一天,在2026年云栖大会上,阿里旗下AI助手千问宣布,将加速打造Personal Agent,为每位用户提供深度理解自己的专属Agent。

今年以来,AI的生产力价值已经得到反复证明;但普通人仍然很难跨过技术这道门槛。但AI越来越强大:什么样的AI才是普通人真正需要的?

Muse、千问们加速布局的Personal Agent方向,刚好适合回答这个问题。

01 Personal Agent强在Personal

从AI应用类型来看,Personal Agent几乎封装了AI的两大专长——陪伴+办事。

Chatbot的长项是「你问我答」,个人助手的角色更多是一个通用的「陪伴者」。而年初爆火的OpenClaw,让人们第一次看到Agent跨应用执行任务的办事能力。

同一时间,刚上线不久的千问也在向「能办事」的AI Agent形态演化,AI购物、AI打车、订机票酒店等能力越来越丰富。

而AI助手面对的是一个具体的人和这个人的长期生活,进入Personal Agent时代是一种必然。

Muse已经把这个方向说得比较明确:它可以基于一次提到的信息,在未来主动调用。比如它记得你上周保存了一份「法式红酒炖牛肉」的食谱,也记得你一个月前在聊天中提到过某位朋友「对麸质过敏」,当你今天说「周末帮我准备一下聚会」时,它会自动把这两件事结合起来。

Muse工程师还展示过更多用法:比如打电话与保险公司讨价还价、取消已经不再使用的流媒体订阅,以及在互联网上寻找折扣。

以此来看,「Personal」有两层含义:一是专属,二是在后台长期存在。理想情况下,一个Personal Agent可以执行一个人持续存在的一组任务和目标。

千问已经在一些领域验证这条路线:在理财场景中,当用户在对话中告诉千问自己的理财目标,千问将结合实时行情、专业机构信息及用户授权连接的持仓数据,提供更贴近个人情况的分析与建议。在健康场景,千问也能知道你孩子多大、体检报告哪几项异常,并能在关键场景提供关键建议。

从用户反馈来看,普通人对这种专属智能的需求,可能远超行业的想象。

Muse下载量280万时,日活用户64万,日活渗透率已经来到23%。而千问在上线10个月后,复杂任务的需求增长超过5倍;用户沟通的轮次来到5-8轮;同时,许多看似简单、简短的提问,也需要有中长期上下文和用户档案画像来确保回答质量更好,这一比例高达62%。

千问产品负责人郑嗣寿强调,个人需求并不等于简单需求,看似简单的问题背后,用户意图千差万别。要满足这些需求,既要能处理复杂任务,更要理解这个具体的人。

国际咨询机构的调研验证了这一点。在凯捷研究院2025年的消费者调查中,已有36%的消费者明确希望获得自动化、个性化的GenAI支持,61%的人希望AI自动补货自己经常购买的商品。埃森哲今年6月发布的另一项调查显示,一旦保障措施到位,80%的人会考虑将至少一项日常任务委托AI代办。

可以说,千问的进化、Muse的上线,刚好顺应了这股日益强烈的专属Agent需求。

02 AI要真正懂一个人有多难?

如果从时间线来看,千问也正恰逢其时地进入了Personal Agent阶段。

上线不到一年,千问总用户数已超过3亿。随着大量真实用户需求进入系统,产品团队逐渐发现一个问题:生活中的不同场景,本来就是相互关联的。运动与健康相关,子女教育会影响理财规划,而一次旅行可能同时涉及消费、交通、住宿和时间安排。

只做好单一场景的Context,反而无法做好这个单一场景。

换句话说,个人AI真正的差距,不在于能处理多少任务,而在于能不能理解基于跨时间、跨场景的连续变化。只有理解「人」的真实生活,AI才能挖掘出更深的需求。

这是因为,个人Context面对的是「人」。一个人可以同时拥有多重身份:企业职员、父母、伴侣、消费者、投资者,也可能正处在求职、育儿、旅行、健康管理等不同人生阶段。

这些身份场景相互渗透——孩子的教育规划可能影响家庭理财,睡眠和健康状态可能影响第二天的工作,而一次工作变动又可能改变居住和消费选择。AI需要在动态中把散落的身份、状态、关系持续拼成一个完整的人。

Personal Agent真正需要建立的,是一个不断更新的「个人状态模型」。

从目前公开的信息来看,千问正在沿着这个方向推进:健康领域,接入智能手表、动态血糖仪等设备,打通个人健康数据;理财领域,接入金融理财类智能体和用户授权的个人持仓数据;围绕个人智能生态,开放平台上的智能体已经扩展到租房、求职、运动、物流、游戏等20多个领域。

这些探索背后,其实对应的是同一个问题:如何把一个人在不同时间、不同场景、不同设备里留下的Context,重新组织成对「这个人」的连续理解?

而千问目前正在从三个维度拓展这套能力:

第一是通用Context,围绕「你是谁」建立动态理解。人的职业、身份、人生阶段和个人偏好都在变化,AI需要在长期互动中提炼这些变化,形成对个人状态的持续理解;

第二是领域级Context,回答「你正在做什么」。如在健康、学习、理财等典型的生活领域中形成专业Context,同时允许不同领域之间调用相关信息;

第三是多终端Context,进一步理解「你此刻正在经历什么」。当手机、手表、PC、AI眼镜以及其他智能硬件产生的信息被连接起来,日常、工作、运动乃至所处环境,都有可能成为Personal Agent理解个体的Context。

过去,一个人的信息散落在不同App、不同服务和不同终端里,彼此之间互不相通;借助Personal Agent,这些孤立的数据被重新组织起来,让AI助手有机会变成一个越来越懂你的AI。

千问在Personal Agent有独特的优势。目前Qwen3.8系列大模型为千问提供了处理复杂Agent任务的模型基础;与此同时,阿里旗下电商、支付、地图、出行等生态服务以及千问开放平台上不断接入的外部服务商,则提供了将「理解」进一步转化为行动的现实接口。

03 AI如何普惠到普通人?

王坚博士在近期分享中提到过一个思考:每一次技术进步,都在重新定义人的能力边界。

随着更高阶的智能的到来,AI的能力边界,也在面向「个人场景」拓展,开始覆盖那些更私人、更复杂、也更难被标准化的部分——诸如如何安排自己的日常生活、如何做选择、如何处理长期目标,以及如何持续表达「我是谁」。

这恰恰是过去的AI助手很难真正进入的领域。因为一个人的生活,很难被拆成一个个独立的Prompt,AI助手要想更好的服务个体,首先需要逐渐形成对一个人的连续理解。

Personal Agent形态的出现,也代表着AI开始从服务具体的任务,走向服务复杂的人。

去年11月上线时,千问就提出过「让智能平等地服务每一个人」的愿景。 如今,这个机会点就落在让没有技术背景的普通人,也可以拥有一个真正属于自己的专属Agent。

Personal Agent时代,不要求用户学习Prompt,不要求用户理解复杂的Agent工作流,它只需要逐渐理解你是谁、你正在经历什么、你在乎什么,并在恰当的时候替你完成一些事情。

「被AI理解」不再只属于极客和企业,每一个普通人都会真正拥有一个方方面面用得上的个人助理。

这或许是Personal Agent真正的价值。

注:文/戴菁,文章来源:降噪NoNoise,本文为作者独立观点,不代表亿邦动力立场。

文章来源:降噪NoNoise

广告
微信
朋友圈

FAQ回顾

Personal Agent是什么?和普通AI助手有什么区别?

Personal Agent是基于用户长期记忆和后台上下文主动提供专属服务的AI智能体,代表如Meta的Muse和阿里千问。它区别于普通Chatbot的“你问我答”,强调Personal(专属)和长期存在,能记住用户过去的偏好和说过的话,并在未来主动调用,执行跨场景的连续任务,例如帮用户筹备聚会、取消订阅或提供个性理财建议。

Meta的Muse为什么能快速成为爆款应用?

Muse是Meta推出的面向普通用户的AI智能体,上线12天下载量达280万,日活用户64万,日活渗透率23%,登上美国App Store榜首。它能基于用户长期信息主动行动,比如记住食谱和朋友过敏信息来安排聚会,或代替用户与保险公司讨价还价,这种“专属且能办事”的体验带动Meta股价上涨。

阿里千问在Personal Agent方向上有哪些布局?

千问宣布加速打造Personal Agent,在理财场景结合实时行情、专业机构信息和用户授权连接的持仓数据提供个性化分析;健康场景接入智能手表、动态血糖仪等设备;开放平台智能体扩展到租房、求职、运动、物流、游戏等20多个领域。千问还从通用Context、领域级Context和多终端Context三个维度构建个人状态模型,以实现对用户的连续理解。

普通人为什么需要Personal Agent?

凯捷研究院2025年消费者调查显示,36%的消费者希望获得自动化、个性化的GenAI支持,61%的人希望AI自动补货常购商品;埃森哲2025年6月调查显示,80%的人愿意在保障措施到位后将至少一项日常任务委托AI。Personal Agent不需要用户学习Prompt或复杂的Agent工作流,只需理解用户是谁、正在经历什么,在恰当时候替用户完成事情,因此对普通人更有价值。

这么好看,分享一下?

朋友圈 分享

APP内打开

赞 +1
+1
微信好友 朋友圈 新浪微博 QQ空间
关闭
收藏成功
发送
/140 0