• 亿邦公众号
    • APP下载

用理解代替推荐,立刻资讯郎许亮,做认知效率的“点火人”

作者:亿邦动力 文章来源:亿邦动力

2026-05-13 16:59

作者:亿邦动力 文章来源:亿邦动力

2026-05-13 16:59

邦小白快读

EN
全文速览

本文介绍了一款解决用户信息痛点的新资讯APP立刻资讯,核心干货如下

1. 解决了传统资讯APP的核心痛点:传统推荐算法只做信息投喂,用户花了大量时间刷内容却没法真正搞懂问题,反而困在信息茧房里,立刻资讯主打用AI提升认知效率,帮用户真正理解问题。

2. 核心功能优势:首创一人一模型,能持续学习摸清你的认知边界,知道你缺什么知识;同一事件提供多身份观点碰撞,带你到观点现场思考;还支持苏格拉底式层层追问,帮你看透事物本质。

3. 对普通用户的价值:能把原来需要两小时整理信息的工作压缩到几分钟,帮你快速理清事件来龙去脉,转化为做决策、做选择的有效依据,大幅提升信息使用效率。

本文透露出内容消费领域的新趋势和用户行为变化,能给品牌商的营销和产品研发提供参考,核心干货如下

1. 消费与用户行为趋势:当前中国内容消费已经从“看内容”转向“理解内容”,78.6%的AI用户使用AI的核心需求是寻找答案、解决问题,用户不再满足于算法投喂的同质化内容,更需要能满足自身深度需求的内容服务。

2. 品牌营销机会:立刻资讯这类新平台用户沉浸度高,单DAU广告价值可达传统资讯产品的数倍,能带来更高的广告溢价,品牌可以提前布局这类新内容平台,获取更好的营销效果。

3. 产品研发启发:品牌做内容营销和用户运营时,需要从原来的流量投喂转向满足用户深度理解需求,帮用户解决实际问题,才能获得更高的用户信任和转化。

本文揭示了内容行业和AI领域的新变化,给各类卖家带来了新机会与风险提示,核心干货如下

1. 行业变化与机会:传统推荐算法赛道已经内卷,行业增长见顶、ARPU值持续下滑,商业模式已经失灵,而AI赋能的深度认知内容是全新的增长赛道,上线半年用户破百万的增速验证了赛道的可行性。

2. 电商卖货新机会:立刻资讯的内容自动化引擎可以快速捕捉热点,自动生成带货脚本、匹配数字人,快速完成从内容生产到带货转化的闭环,能帮助卖家大幅降低内容制作成本,跟上热点节奏提升带货效率。

3. 风险提示:传统依赖低质流量、同质化内容的模式已经不再适配用户需求,卖家需要及时调整方向,可借鉴一人一模型的个性化思路,精准匹配用户需求,提升用户粘性和转化。

本文带来了AI和内容行业的新动向,给工厂推进数字化转型、拓展商业机会提供启示,核心干货如下

1. 产品生产与设计方向:当前用户需求已经从标准化批量供给转向个性化深度匹配,不管是内容还是实体产品,仅仅靠标签定义用户需求已经不够,需要精准捕捉用户未被满足的真实缺口,才能做出适配市场需求的产品。

2. 数字化转型启示:立刻资讯一人一模型持续学习用户认知的思路,值得工厂做数字化转型借鉴,工厂做用户运营和需求收集时,不要只停留在消费标签层面,要深入挖掘用户的真实需求缺口,才能反向指导生产,提升产销匹配效率。

3. 新商业机会:AI内容自动化工具可以帮助工厂做自有品牌电商时,快速生成带货内容,完成从内容到交易的闭环,降低内容生产门槛,帮助工厂拓展线上自有品牌销路,打开新的增长空间。

本文点明了当前内容服务行业的痛点、新趋势和可行解决方案,对各类AI服务商、内容服务商有较高参考价值,核心干货如下

1. 客户痛点与行业趋势:传统推荐算法内容服务只管投喂不管理解,已经无法满足用户需求,当前用户核心痛点是花费大量时间却无法有效解决问题、获得有效认知,78.6%的用户使用AI的核心需求是寻找答案解决问题,这个高阶需求目前没有被行业满足,市场空间广阔。

2. 技术与产品方案参考:立刻资讯推出的一人一模型+多Agent观点群+苏格拉底式互动的三层架构,是解决用户深度认知需求的可行路径,核心是抓住用户认知缺口,通过观点碰撞和持续追问帮助用户真正理解内容,可作为服务商开发产品的核心方向。

3. 商业化方向参考:除广告变现外,认知模型能力付费,切入高净值决策场景,是爆发力更强的变现路径,且用户转换成本极高,能构建深厚护城河。

本文介绍了资讯内容平台的创新方向,对平台的运营、招商和风险规避都有参考价值,核心干货如下

1. 用户对内容平台的新需求:当前用户已经不再满足于传统推荐算法打造的信息茧房,更多用户是带着问题来获取信息,需要平台能帮自己真正理解内容、解决问题,传统平台如果不调整,会持续面临用户流失、增长见顶的问题。

2. 平台运营创新参考:可以借鉴立刻资讯的运营和产品逻辑,打造适配用户个性化认知的一人一模型系统,通过多观点碰撞提升用户深度思考意愿,再通过可追问的互动提升用户沉浸度,能有效提升用户停留时长和粘性,拉高平台广告价值。

3. 招商与风险提示:AI认知引擎赛道增长速度快,立刻资讯上线半年用户破百万,且具备清晰的商业化路径,具备较高的招商和投资价值;平台需要注意规避传统推荐算法内卷的风险,提前布局认知服务新赛道。

本文展现了AI赋能内容产业的全新动向,提出了新的技术路径和商业模式,对产业研究有较高的参考价值,核心干货如下

1. 产业新动向:当前内容产业已经进入新的发展阶段,过去十年的推荐算法模式已经走到内卷,行业整体面临增长见顶、用户流失、ARPU值下滑、商业模式失灵的问题,内容消费从看内容转向理解内容,GenAI赋能的深度认知服务成为全新的产业方向。

2. 技术创新:立刻资讯全球首创一人一模型技术,区别于传统标签式用户画像,能持续学习感知用户的认知边界和需求缺口,搭配多Agent观点群、苏格拉底式持续下钻互动,形成了全新的产品技术架构,构建了差异化壁垒。

3. 商业模式创新:提出了高溢价广告+认知模型能力付费的双路径商业模式,认知付费依托和用户认知共生的AI分身,形成了极高的用户转换成本,构建了难以复制的终极护城河,未来还可拓展多内容形态、电商带货、全球化布局,打开了全新的商业想象空间。

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

This article introduces Lilike Zixun (Instant News), a new news app built to address key pain points in information consumption. Its core value propositions are as follows:

1. It solves a core problem of traditional news apps: conventional recommendation algorithms only push feed content, leaving users stuck in information cocoons after spending hours scrolling without actually understanding the issues they care about. Lilike Zixun leverages AI to improve cognitive efficiency, and helps users develop a genuine understanding of topics.

2. Core functional advantages: It is the first app to adopt a "one-model-per-user" design, which continuously learns to map the user’s cognitive boundaries and identifies knowledge gaps. It also presents multiple perspectives from different stakeholder identities for any given event to immerse users in on-the-ground critical thinking, and supports Socratic layered questioning to help users get to the root of any issue.

3. Value for general users: It compresses information organization work that would normally take two hours into just a few minutes, helping users quickly sort out the context of events, generate actionable insights for decision-making, and drastically improve information utilization efficiency.

This article outlines emerging trends and shifting user behaviors in content consumption, offering key insights for brand marketing and product R&D. Core takeaways are as follows:

1. Trends in consumer and user behavior: Content consumption in China has shifted from "consuming content" to "understanding content". 78.6% of AI users use the technology primarily to find answers and solve problems. Users are no longer satisfied with homogeneous content pushed by algorithms, and increasingly demand content services that meet their deep, personalized needs.

2. Brand marketing opportunities: New platforms like Lilike Zixun achieve far higher user engagement. The ad value per daily active user (DAU) can be multiple times that of traditional news products, delivering higher ad premiums. Brands can get ahead of the curve by establishing an early presence on these new content platforms to unlock stronger marketing performance.

3. Insights for product R&D: When developing content marketing strategies and user operations, brands need to shift from the old model of traffic feeding to prioritizing users’ demand for deep understanding and helping users solve practical problems. This approach is the only way to build higher user trust and drive stronger conversion.

This article reveals new shifts in the content and AI industries, outlining emerging opportunities and risk alerts for sellers of all types. Core takeaways are as follows:

1. Industry shifts and opportunities: The traditional recommendation algorithm track is already hyper-competitive, with stagnant industry growth, steadily declining ARPU, and broken business models. AI-powered deep cognitive content represents an entirely new growth track, and its feasibility is confirmed by Lilike Zixun hitting 1 million registered users within six months of launch.

2. New opportunities for e-commerce sales: Lilike Zixun’s automated content engine can quickly capture trending topics, automatically generate sales scripts, match them with digital avatars, and close the loop from content creation to conversion. This helps sellers drastically cut content production costs, keep up with trending topics, and improve sales efficiency.

3. Risk alerts: The traditional model that relies on low-quality traffic and homogeneous content no longer aligns with user demand. Sellers need to adjust their strategy in a timely manner, and can adopt Lilike Zixun’s "one-model-per-user" personalization approach to match user demand accurately and boost user retention and conversion.

This article covers new developments in AI and the content industry, offering insights for factories pursuing digital transformation and expanding business opportunities. Core takeaways are as follows:

1. Product development and design direction: User demand has shifted from standardized mass supply to personalized, depth-aligned offerings. For both content and physical products, defining user needs via tags alone is no longer sufficient. Businesses must accurately capture unmet real user needs to build products that fit market demand.

2. Insights for digital transformation: Lilike Zixun’s "one-model-per-user" approach, which continuously learns user cognitive profiles, offers a useful framework for factories undergoing digital transformation. When collecting user data and managing user relationships, factories should go beyond surface-level consumption tags to dig into users’ real unmet needs. This allows factories to use user insights to guide production and improve the efficiency of demand-supply matching.

3. New business opportunities: AI-powered automated content tools can help factories build their own D2C e-commerce brands by quickly generating sales content, closing the loop from content to transaction, lowering barriers to content production, helping expand online sales of owned brands, and opening up new growth avenues.

This article identifies core pain points, emerging trends, and actionable solutions in the content service industry, offering high-value reference for AI and content service providers of all types. Core takeaways are as follows:

1. Client pain points and industry trends: Traditional recommendation algorithm-based content services only focus on content feeding, not user understanding, and can no longer meet user demand. Today’s core user pain point is that users spend massive amounts of time but fail to effectively solve problems or gain useful cognition. 78.6% of AI users use the technology primarily to find answers and solve problems, and this high-level demand remains largely unmet by the industry, leaving significant untapped market opportunity.

2. Reference for technical and product roadmaps: Lilike Zixun’s three-layer architecture — one model per user, multi-agent perspective group, and Socratic interactive questioning — represents a viable path to addressing users’ deep cognitive demand. The core of this approach is centering on users’ individual cognitive gaps, helping them truly understand content via perspective sharing and continuous questioning. This can serve as a core direction for service providers developing new products.

3. Reference for commercialization: Beyond advertising monetization, monetizing cognitive model capabilities through paid access for high-net-worth decision-making scenarios is a more high-growth monetization path. It also creates extremely high user switching costs, allowing businesses to build deep competitive moats.

This article outlines the innovation direction for news content platforms, offering reference for platform operations, recruitment, and risk mitigation. Core takeaways are as follows:

1. New user demand for content platforms: Users are no longer satisfied with the information cocoons created by traditional recommendation algorithms. More users come to content platforms with specific problems to solve, and need platforms to help them truly understand content and resolve issues. Traditional platforms that fail to adapt will continue to face user attrition and stagnant growth.

2. Reference for operational innovation: Platforms can adopt Lilike Zixun’s product and operational logic to build a "one-model-per-user" system tailored to users’ individual cognitive profiles. Multiple competing perspectives deepen users’ willingness to think critically, while interactive questioning boosts engagement. This framework effectively improves user session length, retention, and overall platform ad value.

3. Investment and recruitment insights and risk alerts: The AI cognitive engine track is growing rapidly. Lilike Zixun gained 1 million users in six months of launch with a clear commercialization path, making it a high-value opportunity for recruitment and investment. Platforms should avoid the risk of hyper-competition in the traditional recommendation algorithm space, and move early to enter the new cognitive services track.

This article presents new developments in AI-powered content industry transformation, and proposes a new technical path and business model, offering high reference value for industry research. Core findings are as follows:

1. New industry developments: The content industry has entered a new development stage. The recommendation algorithm model that dominated the past decade has become hyper-competitive, with the broader industry facing stagnant growth, user attrition, declining ARPU, and broken business models. Content consumption has shifted from content consumption to content understanding, and GenAI-powered deep cognitive services have emerged as an entirely new industry direction.

2. Technological innovation: Lilike Zixun is the first in the world to launch the "one-model-per-user" technology. Unlike traditional tag-based user profiling, this model continuously learns to map the user’s cognitive boundaries and unmet needs. Combined with a multi-agent perspective group and Socratic layered interactive questioning, it forms an entirely new product and technology architecture that creates strong differentiated barriers to entry.

3. Business model innovation: It introduces a dual-path business model of high-premium advertising plus paid cognitive model capabilities. Paid cognition is built on an AI co-evolutionary "cognitive twin" that aligns with the user’s evolving knowledge, creating extremely high user switching costs and building a hard-to-replicate ultimate competitive moat. In the future, this model can be extended to support multiple content formats, e-commerce sales, and global expansion, opening up entirely new commercial opportunities.

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.

当中国内容消费从“看内容”转向“理解内容”,一个以GenAI深度理解为核心的APP正浮出水面。

立刻资讯,是北京智讯蜂巢信息服务有限公司的战略级核心产品,获领润生态重点孵化。上线六个月,用户破百万。在综合资讯App格局已定的今天,这个增速验证了一个极具商业价值的反向假设:用理解代替推荐的时代已经到来。

立刻资讯总裁郎许亮

如果说过去十年,移动互联网用推荐算法给每个人建了一座信息茧房,那郎许亮正试图用AI在这座茧房上凿开一扇窗。

这个清华系毕业的创业者,曾在字节跳动、一点资讯主导过年收入过十亿的产品,见证了推荐算法从鼎盛走向内卷。2025年,当整个资讯行业困在增长见顶、用户流失、商业模式失效的泥潭里时,他带着“立刻资讯”出场了。

茧房破裂:从“信息投喂”到“认知效率跃迁”

当下,推荐算法一个致命的行业痛点是,只管内容投喂,不管内容理解。 用户花的时间越来越多,真正搞懂的事却越来越少。中国网民规模已突破11.3亿,互联网普及率超80%,综合资讯类App的单机单日使用时长从2024年的89分钟涨到了2025年的98分钟。但悖论是“流量”不等于“留量”,时长不等于变现效率,行业ARPU值反而在见底。推荐算法的商业模式正在“失灵”。

问题出在哪?郎许亮有一个很形象的比喻:“这就好比你急需一把钥匙开锁,算法却给你推了一万把看着很像的钥匙。你试到天黑也打不开门,最后只收获了一堆废铁。” 信息看似海量,但对于用户想搞懂的那个问题而言,绝大多数都是无效的噪音。

更深的证据藏在AI的使用数据里。当被问及“用AI主要做什么”时,78.6%的用户回答是“寻找答案,解决问题”,76%是“生成和处理文本”。这意味着什么?意味着用户打开手机,根本不是来“刷”内容的,他是带着问题来的,他想“弄明白”一件事。

可打开主流资讯App,扑面而来的还是“猜你喜欢”。“你喜欢什么它猜得挺准,但你想搞懂什么,它完全不管。”郎许亮说,“这是典型的拿着金饭碗要饭——用户的高阶需求就摆在那儿,你接不住。”

立刻资讯要接的,就是这个需求。在郎许亮的定义里,立刻资讯的对手从来不是今日头条。它的使命是终结用户在各个App之间疲于奔命地搜索、查证、比对观点的无效消耗,用AI实现“认知效率”的根本性提升。

什么是认知效率?把两小时“拼图”,压缩成五分钟的决策武器

这个洞察,来自郎许亮上一段产品生涯里挥之不去的观察。

他身边一位做投资的朋友,每次出了财经大事件,都要同时打开四个App。看快讯、看深度、扒情绪、盯行情。一圈下来,两个小时没了。问他看懂了没有,他说:懵懵懂懂,好像知道了点什么。

“好像知道了点什么”——这七个字,是整个行业最大的失败。信息给了,但理解没有发生;用户吃进去了,但没消化。

这句话,成了立刻资讯的哲学原点:上一个十年,推荐算法让所有人吃得饱;下一个十年,AI得教会每个人怎么消化。

立刻资讯要做的,是一个能把那两小时信息拼图,压缩成两分钟逻辑快照,AI不仅告诉你发生了什么,还要把事件的来龙去脉、多方观点、因果链条理清,甚至追着你问问题,逼着你思考。

但理解只是第一步,真正的认知效率,是让信息为你所用。在立刻资讯的逻辑里,世界上发生的每一个大事件、每一轮政策变动、每一场行业风暴,都不应仅仅是茶余饭后的谈资,这些内容必须成为你的利好武器。

当你真正弄懂了一个事件的因果,你就能在投资决策中抢占先机,在职业选择上避开雷区,在商业博弈中看透底牌。从“看热闹”到“看门道”,再把门道转化为你的行动势能,这才是认知效率的完整闭环,也是立刻资讯要做的事情。

降维打击:全球首创“一人一模型”

从根上讲,立刻资讯做的不是一款“资讯App”,而是“深度理解的认知引擎”。这不仅是个产品定位,更是决定护城河深浅的设计逻辑。撑起这一逻辑的核心壁垒,是立刻资讯全球首创的“一人一模型”。

传统用户画像用标签定义人:金融男、高收入、爱滑雪。但标签永远回答不了一个致命问题:这个金融男在“芯片制裁”上的认知水平是什么?他到底想搞懂什么?标签知道你消费了什么,却永远不知道你不懂什么。

立刻资讯的“一人一模型”,是一个会持续学习的AI认知分身。用户问得越深,它越懂你的认知边界。它不猜你爱看什么,而是感知你还缺什么。推荐算法讨好你的欲望,一人一模型触碰你的无知——这是两个时代的分水岭。

读懂了人,还要重构人所处的信息场。于是有了第二层,也是最狠的一招:多Agent观点群。

所有AI解读新闻都是同一个声音、一个结论。但真相从来不在单一视角里产生,真理来自观点的冲突。立刻资讯打造了一座“观点的罗马斗兽场”:同一事件,乐观派专家、数据狂人、理性分析者、阴谋论专家,甚至吃瓜群众,在同一张桌子上公开博弈。用户在这些声音的撕扯中被迫开始思考。

产品负责人李骏把一句话刻在团队墙上:“这不是情绪陪伴,这是把用户拉到观点的现场。你不是观众,你是陪审团。”

观点碰撞激发出思考,但思考不能悬停在表面。立刻资讯随即递上第三重:“苏格拉底式互动”持续下钻。

传统的阅读是单向接收,看完就结束。但在立刻资讯,基于“一人一模型”的支撑,用户可以顺着自己的逻辑盲区,对任何一条信息进行持续追问和下钻,“背后的核心数据是什么?” “如果这个变量变了,结果会怎样?”AI不会给你一个干瘪的结论,而是陪你层层剥笋,直到你真正看透事物的本质。

这一切,最终都指向同一个信条:提问胜过灌输,推理胜过记忆,思考胜过阅读。

商业重构:从流量变现到“认知付费”

产品立住了,商业化就是必答题。立刻资讯的路径很清晰:两条腿走路,且后一条比前一条更具爆发力。

第一条腿,高溢价广告变现。这看起来是资讯产品的常规动作,但底层逻辑已截然不同。传统资讯是“流量洗稿”,用户匆匆划过;立刻资讯则是“深度停留”,用户在结构化思考和深度互动中沉浸更久。这种高浓度的注意力,让单DAU的广告价值可达传统资讯的数倍。

然而真正令资本市场兴奋的,是第二条腿——模型能力变现,也就是“第二曲线”。一旦AI深度理解了一个人的认知,立刻资讯便能介入他所有高净值决策场景。用户过去花两小时研读报告才能得出的结论,现在几分钟就能拿到。这种硬核的决策赋能,天然具备极强的付费意愿。更关键的是,它不是一次性工具,而是越用越懂你的认知共生体:它持续记忆你的信息视角、思维偏好,甚至能预判你在哪个逻辑节点容易卡住。到那时,用户想换都换不掉,因为这个AI已长成他认知结构的一部分。在郎许亮看来,这才是终极护城河:“标签可以被复制,算法可以被模仿,但一个和你认知共生了几百个小时的AI分身,其转换成本无人能跨越。”

望向未来,其蓝图远不止一款资讯App。内容自动化引擎正将边界推向AI视频、播客、数字人乃至电商带货。热点被捕捉后,瞬间转写脚本、生成口播、匹配数字人、变身带货内容,“被看到、被听到、被买到”的闭环一旦形成,商业想象空间就会彻底引爆。

更远处,是全球化的战略布局。立刻资讯的海外版即将上线。当语言不再是障碍,AI要解决的是思维的转译。同一事件,不同文明背景的用户追问方向截然不同,而立刻资讯的“一人一模型”有能力捕捉这种思维差异,这不是把资讯推向全球,这是让“通过理解而认知”这件事,成为人类共通的信息权利。

走廊里,技术团队正在调试多Agent对话模型,屏幕上几个AI正为一个政策事件的解读吵得不可开交。

有人问他,做“理解”这件事,觉得难吗?

他笑了笑,像是对来客,也像是对自己说:“难走的路从不拥挤”

然后他转身,扎回了那个由数据、模型和追问构成的AI世界。在他身后,是一个被推荐算法投喂了整整一代人的信息洪流;而他要做的,是为这条洪流装上一道闸门,教会每一个身处其中的人,如何真正消化——这不仅是产品的胜利,更是“认知效率,理解效率”的复利。

【版权提示】亿邦动力倡导尊重与保护知识产权。未经许可,任何人不得复制、转载、或以其他方式使用本网站的内容。如发现本站文章存在版权问题,烦请提供版权疑问、身份证明、版权证明、联系方式等发邮件至run@ebrun.com,我们将及时沟通与处理。

已点赞

+1

{{ praise_text }}

好文章,值得被更多人看到

FAQ回顾

什么是立刻资讯的核心产品定位?

立刻资讯的核心定位不是传统资讯App,而是一个“深度理解的认知引擎”。它旨在通过AI技术,终结用户在多个App间搜索、查证信息的无效消耗,从根本上提升用户的“认知效率”。

立刻资讯的“一人一模型”技术如何运作?

“一人一模型”是全球首创的AI认知分身技术,它持续学习用户的认知边界,感知用户不懂什么,而非猜测用户喜欢什么。它通过苏格拉底式互动,允许用户对信息进行持续追问和下钻,帮助用户看透事物本质。

立刻资讯的商业化模式有哪些?

立刻资讯采用两条商业化路径:一是基于深度停留的高溢价广告变现;二是更具潜力的模型能力变现(第二曲线),即通过AI深度理解赋能用户的高净值决策场景,实现“认知付费”的订阅或服务模式。

广告

下一篇

跨境电商资源对接

《全球跨境电商服务商资源手册》第二期手册现已发布,触达更多资源,了解跨境服务商信息,欢迎点击链接下载。

了解详情

科研破局、引领行业,汤臣倍健:做良性生态的构建者

汤臣倍健作为VDS行业领导品牌,面对行业同质化内卷与信任危机,选择以长期主义破局:通过牵头制定行业标准