广告
加载中

对话马啸:硬件将成为AI理解世界的“眼睛”和“耳朵” 未来智能的可能是让AI“隐形”

王亚静 2026-05-25 09:18
王亚静 2026/05/25 09:18

邦小白快读

EN
全文速览

本文是AI耳机创业公司未来智能CEO马啸的专访,整理了AI硬件行业的核心信息和干货内容。

1. 核心行业观点:当前AI+硬件的清晰方向是让硬件成为AI获取外界信息的“眼睛”和“耳朵”,好的AI产品应当主动适应用户习惯,做到“AI隐形”,让用户感受不到AI的存在,才能覆盖大多数普通用户。

2. 新品核心优势:未来智能最新发布的viaim讯飞智能体耳机,升级了AI Agent能力,可以围绕同一主题处理完整上下文信息,自动帮办公用户整理待办、跟进未完成事项、整合经营数据,相当于随身办公小助手。

3. 创业感悟参考:创业者要坚持“进窄门,走宽路”,深耕垂直场景,不急不躁做好产品基本盘,不要盲目蹭风口冒进。

本文给AI硬件领域的品牌商提供了行业趋势、产品研发、渠道合作等多方面的干货参考。

1. 消费趋势洞察:用户对AI硬件的需求已经从单一功能转变成全流程任务处理,用户需要AI主动适配自身习惯,降低使用门槛。当前AI仅被1%的精英用户使用,降门槛让AI隐形才能打开大众市场。

2. 产品研发方向:要避开随便加AI大模型蹭概念的行业虚火,聚焦垂直场景深挖需求,未来智能坚持死磕办公会议场景,走“进窄门走宽路”的路线,新品升级成AI Agent工作系统,契合用户对全流程信息处理的需求。

3. 渠道合作思路:融资和合作优先选择能产生业务协同的对象,可以借助投资方的生态和渠道资源拓展市场,比如对接蚂蚁的国内生态、传音的海外分销体系实现共赢。

本文给布局AI硬件赛道的卖家梳理了市场机会、风险提示和可借鉴的发展经验。

1. 市场增长机会:AI+硬件已经告别早期概念探索阶段,行业方向逐渐清晰,硬件作为AI信息入口的价值明确,办公垂直赛道的AI硬件仍有较大增长空间,海外新兴市场增速极快,未来智能2026年一季度海外全渠道增长3.8倍,活跃用户增长11.5倍,是值得挖掘的增量市场。

2. 风险提示:早期不少玩家只是在硬件中硬加AI大模型,没有落地解决用户真实需求,最终都走向失败,卖家要避开蹭风口做概念产品的陷阱,不要为了热点推出没有实际价值的产品。

3. 可借鉴的合作模式:可以对接拥有生态或渠道资源的合作方,借助对方的资源触达更多用户,打开国内外市场,降低自身拓展成本。

本文给布局AI硬件赛道的工厂分享了产品需求、商业机会和数字化发展启示。

1. 产品生产设计需求:AI硬件不能只拼硬件参数,还要重视软件AI能力的匹配,要先把硬件性能做扎实,再升级AI能力适配用户需求。当前用户需要的不是零散的功能,而是能承接全流程任务的系统,产品设计要围绕完整任务链路展开,不能只是简单堆砌AI概念。

2. 商业合作机会:办公垂直场景的AI硬件还有很大拓展空间,未来智能已经积累超150万用户,接下来还会推出更多办公场景新形态产品,会给上游工厂带来稳定的订单需求。同时海外AI硬件市场增长迅猛,也会催生更多出海相关的生产合作机会。

3. 发展启示:工厂布局AI硬件要避开盲目蹭风口的误区,聚焦垂直场景深耕,夯实产品基本盘,不要冒进扩张。

本文给AI硬件相关服务商梳理了行业发展趋势、客户痛点和可布局的业务方向。

1. 行业发展趋势:AI+硬件已经从早期概念探索走向落地阶段,行业逐步形成共识,硬件的核心定位就是AI获取上下文信息的入口,AI Agent与硬件结合是明确的发展方向,产品范式升级会带来大量配套服务需求。

2. 客户核心痛点:行业目前普遍存在两类痛点,一是不少中小AI硬件厂商缺乏AI技术开发能力,难以满足用户对全流程任务处理的真实需求,只能做概念产品难落地;二是多数厂商缺乏生态和渠道资源,难以触达C端用户打开市场,尤其是出海渠道资源稀缺。

3. 业务布局方向:服务商可以围绕垂直场景,为厂商提供AI Agent能力开发、全流程任务系统搭建的技术服务,也可以做生态对接服务商,为厂商对接国内生态资源和海外分销渠道。

本文给布局AI硬件赛道的平台商梳理了行业需求、招商方向和风险规避思路。

1. 行业对平台的核心需求:AI硬件创业品牌,尤其是深耕垂直赛道的品牌,非常需要平台提供生态场景接入、分销渠道资源支持,不管是国内的消费场景接入,还是海外新兴市场的渠道对接,都是品牌迫切需要的资源。

2. 平台招商布局方向:可以优先引入深耕垂直场景的AI硬件品牌,当前AI办公硬件是增长极快的细分赛道,用户付费意愿很高,未来智能订阅用户同比增长高达23.8倍,能给平台带来新的增量。同时良性竞争可以共同培育新市场,平台可以引入不同品牌把赛道蛋糕做大。

3. 风向规避思路:要避开概念化的AI硬件项目,优先选择聚焦真实用户需求、有成熟用户基础、深耕场景的品牌,不要盲目蹭风口引入没有实际价值的项目,避免风口退去后项目失败给平台带来损失。

本文给AI产业研究者提供了AI+硬件领域最新的产业动向、创业案例和行业思考,参考价值很高。

1. 最新产业动向:AI+硬件经过多年探索后,发展方向逐渐清晰,硬件的核心定位是AI获取上下文信息的“眼睛”和“耳朵”,AI Agent与硬件结合是新的产品范式,垂直深耕成为行业先行者的共同选择,AI办公硬件已经跑出了成熟项目,用户规模突破150万,遍布全球200多个国家和地区,海外市场增长迅猛。

2. 行业待解决的新问题:当前AI行业存在使用门槛过高的问题,仅有1%的精英用户会主动使用AI,如何降低门槛让AI隐形适应用户习惯,是全行业需要解决的核心问题,早期大量概念化AI硬件因为缺乏落地场景大面积失败,也值得研究者总结反思。

3. 商业模式参考:“进窄门走宽路”的垂直深耕模式,以及优先选择业务协同投资方的融资策略,都为AI创业领域的研究提供了鲜活的案例。

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

This article is an exclusive interview with Ma Xiao, CEO of AI headphone startup Yunzhitang (Future Intelligence). It compiles core insights and practical takeaways from the AI hardware industry.

1. Core industry perspective: The clearest current direction for AI + hardware is to position hardware as the "eyes and ears" through which AI captures external information. A good AI product should proactively adapt to user habits and achieve "invisible AI"—users should not feel the presence of AI at all, making it accessible to most mainstream consumers.

2. Core advantages of the new product: Future Intelligence's newly launched Viaim iFlytek AI Agent Headphone has upgraded AI Agent capabilities, enabling it to process complete contextual information around a single topic. It automatically organizes to-dos for office users, follows up on unfinished tasks, and integrates business data, acting as a portable personal office assistant.

3. Lessons for entrepreneurs: Founders should stick to the principle of "entering a narrow door to walk a wide road": focus deeply on vertical scenarios, build a solid product foundation steadily, and avoid blindly chasing trends to rush into expansion.

This article provides practical insights on industry trends, product development and channel cooperation for brands in the AI hardware space.

1. Consumer trend insight: User demand for AI hardware has shifted from standalone features to end-to-end task processing. Users expect AI to adapt proactively to their habits and lower the barrier to use. Currently, AI is only used by 1% of power users. Lowering access barriers and making AI invisible is the only way to unlock the mass market.

2. Product R&D direction: Brands should avoid the industry fad of simply adding a large AI model for marketing hype, and instead dig deep into needs in focused vertical scenarios. Future Intelligence has taken the "narrow door to wide road" approach by focusing relentlessly on office and meeting scenarios. Its new product has been upgraded to an AI Agent work system that aligns with user demand for full-process information processing.

3. Channel and partnership strategy: Prioritize investors and partners that can deliver business synergy. Brands can leverage their partners' ecosystem and channel resources to expand into new markets. For example, Future Intelligence has partnered to access Ant Group's domestic ecosystem and Transsion's overseas distribution network to achieve mutual growth.

This article outlines market opportunities, risk warnings and actionable development lessons for sellers entering the AI hardware track.

1. Growth opportunities: AI + hardware has moved past the early concept exploration phase, with a clearer industry direction and a confirmed value for hardware as an AI information entry point. AI hardware for the vertical office vertical still holds significant growth potential, and emerging overseas markets are growing extremely rapidly. Future Intelligence recorded 3.8x full-channel overseas growth and 11.5x active user growth in Q1 2026, making this a high-potential incremental market worth tapping into.

2. Risk warning: Many early players simply forced a large AI model into their hardware without solving actual user needs, and ultimately failed. Sellers should avoid the trap of creating concept products just to chase trends, and should not launch valueless products just to capitalize on hype.

3. Replicable partnership models: Sellers can partner with players that own ecosystem or channel resources, leverage their resources to reach more users and open up domestic and overseas markets, while reducing their own expansion costs.

This article shares insights on product demand, business opportunities and digital development for factories entering the AI hardware track.

1. Product design and manufacturing requirements: AI hardware cannot compete on hardware specs alone; manufacturers must also prioritize matching software and AI capabilities. Factories should first solidify core hardware performance, then upgrade AI capabilities to meet user needs. Today's users don't want scattered features—they want systems that can handle end-to-end tasks, so product design must be built around complete task workflows, rather than simply stacking AI concepts.

2. Business cooperation opportunities: AI hardware for office vertical scenarios still has large room for expansion. Future Intelligence has already accumulated more than 1.5 million users, and will launch more new form-factor products for office scenarios, bringing stable order demand to upstream factories. The fast-growing overseas AI hardware market will also create more production cooperation opportunities for出海-focused manufacturers.

3. Development insights: Factories entering AI hardware should avoid the mistake of blindly chasing hype, focus deeply on vertical scenarios, build a solid product foundation, and avoid reckless expansion.

This article outlines industry development trends, client pain points and actionable business directions for service providers serving the AI hardware industry.

1. Industry development trends: AI + hardware has progressed from early concept exploration to real-world deployment, and industry consensus is gradually forming: the core positioning of hardware is as an entry point for AI to capture contextual information, and the combination of AI Agent and hardware is a clear development direction. This product paradigm shift will create massive demand for supporting services.

2. Core client pain points: The industry currently faces two widespread pain points. First, many small and mid-sized AI hardware brands lack AI development capabilities, and cannot meet user demand for end-to-end task processing—they can only build concept products that fail to deliver real value. Second, most brands lack ecosystem and channel resources to reach end consumers and scale, especially when it comes to overseas distribution resources.

3. Business expansion directions: Service providers can focus on vertical scenarios, offering AI Agent development and end-to-end task system building technical services for hardware brands. They can also act as ecosystem connection service providers, helping brands access domestic ecosystem resources and overseas distribution channels.

This article outlines industry demand,招商 priorities and risk mitigation strategies for platforms entering the AI hardware track.

1. Core industry demand from platforms: AI hardware startups, especially those focused on vertical tracks, have strong demand for platforms to provide ecosystem access and distribution channel support. Both access to domestic consumption scenarios and connections to channels in emerging overseas markets are resources that brands urgently need.

2.招商 and positioning direction for platforms: Platforms should prioritize onboarding AI hardware brands that focus deeply on vertical scenarios. AI office hardware is currently one of the fastest-growing niche tracks, with very high user willingness to pay. Future Intelligence's subscription user base grew 23.8x year-over-year, which can bring new incremental traffic and revenue to platforms. Healthy competition can also help co-develop the new market, so platforms can onboard multiple brands to grow the overall size of the track.

3. Risk mitigation strategies: Platforms should avoid onboarding purely conceptual AI hardware projects, and prioritize brands that focus on real user needs, have an established user base, and focus deeply on specific scenarios. Don't onboard low-value projects just to chase hype, which can leave platforms with losses when the trend fades and projects fail.

This article provides the latest industry developments, startup cases and industry thinking for AI industry researchers focused on AI + hardware, with high reference value.

1. Latest industry developments: After years of exploration, the development direction of AI + hardware is gradually clearing up. Hardware's core positioning is as the "eyes and ears" for AI to capture contextual information. The combination of AI Agent and hardware represents the new product paradigm, and deep vertical focus has become the common choice for industry first movers. Mature players have emerged in the AI office hardware space: one leading player, Future Intelligence, has surpassed 1.5 million global users across more than 200 countries and regions, with extremely fast growth in overseas markets.

2. Unresolved core industry issues: The current AI industry faces a major problem of overly high barriers to entry. Only 1% of power users actively use AI today. How to lower barriers and make invisible AI that adapts to user habits is the core problem the entire industry needs to solve. The large-scale failure of early purely conceptual AI hardware projects due to lack of real deployment scenarios is also an important topic for researchers to analyze and reflect on.

3. Reference for business model research: The deep vertical focus model of "entering a narrow door to walk a wide road", as well as the financing strategy of prioritizing investors that deliver business synergy, both provide fresh, real-world cases for research on AI entrepreneurship.

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.

带领未来智能扎根AI耳机5年,马啸更从容了。

5月20日,未来智能在北京举行了一个相对小型的发布会,作为CEO的马啸带着两款「viaim讯飞智能体耳机」新品现身。和去年亮相新品发布会时相比,马啸瘦了、笑容多了,而且更轻松了。

马啸有从容的底气。

自2021年从科大讯飞独立以来,未来智能就踏入AI耳机赛道,死磕办公会议场景,如今用户已经超过150万;在举行发布会前两周,公司刚刚完成亿元级融资,传音参投,销售、融资齐头并进。

发布会结束后,马啸接受了「创业最前线」的专访,讲述了未来智能在AI耳机上的布局与战略,以及对融资、上市的思考。在采访中,他毫不讳言市场竞争压力之大,却坚定信念——绝不做“镜花水月”之事,只想脚踏实地做实事。

“进窄门,走宽路”,马啸用这句话概括未来智能的产品哲学,也用这句话表达自己对公司的期待。

01 “最好感受不到AI的存在”

如果在5、6年前谈起AI,人们只会觉得那是遥不可及的畅想。但自从ChatGPT出现之后,AI就随着冰箱电视、耳机、眼镜等载体猝不及防地闯进我们的生活。

2022年,AI大模型集中涌现;到了2024年,华强北已经可以推出各式各样的智能硬件。马啸笑谈:那个阶段如果问“AI和硬件结合起来能做什么”,可能100个公司有100个答案。

那个时候,市场上的想法天马行空,好像硬件里随便加一个AI大模型就可以被叫做AI硬件——耳机里塞一个ChatGPT叫AI耳机,胸口挂一个能录音的挂坠就叫AI挂坠,但很多产品始终难以真正落地推广,最终都失败了。

到了2026年,AI Agent产品小龙虾突然爆火,这让“AI+硬件=?”这个问题的答案越来越清晰。马啸发现,绝大多数To C硬件公司都在做同一件事——让硬件成为AI理解世界、获取context(上下文)的“眼睛”和“耳朵”。

而这,是未来智能从2021年起就在做的事情。

“原先的思维还是不够先进。”尽管认知上提早了5年,但马啸仍在反思,过去的产品逻辑是把AI当作一个闭环工具——做记录、转写、翻译,任务到此结束。

但他发现,用户的真实行为是把录音转成文字后,又丢到大模型里或自己的记录软件里继续处理。“用户现在对AI的需求绝不仅仅是记录,而是需完成整个任务。我们去年还觉得用户不愿意留在我们这儿,是因为我们没有后续的服务,当时也不清楚如何提供这种服务。”

直到小龙虾爆火,马啸被震动了,仿佛看到很多无形中的“墙”被打通了,“我们为什么要守着锁死功能,把内容留给别人,最后留给龙虾去处理,我们自己为什么不是龙虾呢?”

在他看来,现在AI是1%人的狂欢和99%人的懵圈,小龙虾让大家对AI Agent认知群体变大了,但使用门槛依然没有降下来,“如果要更广泛的被使用,AI就得更加主动。”

从2021年创业起,马啸的理念就是AI耳机不在于AI炫技,而是融入用户的行为习惯。正因为此,未来智能对AI耳机的迭代方向是,让用户在使用AI耳机的过程中“感受不到在使用AI”。

简单来说,当足够的文件传输到viaim以后,AI会直接提出操作请求,用户只需要“打勾打叉”,“大模型把各种各样信息尽可能掌握了,然后直接就完成这个任务。”

“让用户不要起心动念。”马啸认为只有AI主动了,99%的用户才有可能会去用,“AI不主动,它在那儿高高在上,只有1%的精英用户去用。”

他相信,真正好的AI产品不是让用户去学习AI,而是让AI去适应人的习惯,当AI足够隐形的时候,它才真正走进了用户的心里。

02 AI耳机,不只是耳机

2025年,未来智能邀请调音团队进行专业调校,对音质进行全面升级,是为了让硬件更“硬”。

到了2026年的新品发布会上,马啸带着全新升级的viaim讯飞智能体耳机亮相。这一次,他想让用户感受到AI耳机的“软”。

据了解,未来智能此次推出的viaim讯飞智能体耳机首次引入了“项目”“技能”“开放输入”等多项AI Agent能力,使AI能够理解同一主题下的持续上下文、按预设方法论主动处理内容、并支持外部音频与文档的输入。

自此,AI不再只理解一条孤立的录音,而是能够理解围绕同一主题持续积累的完整上下文,从“处理一次内容”升级为“推进一件事情”。

“这不是一个新功能,这是一种新范式。”马啸在发布会现场这样定义这次升级。在他看来,用户真正需要的不是更多分散功能,而是一套能持续承接信息、处理信息并产出结果的工作系统。

在马啸看来,viaim讯飞智能体耳机的诞生是未来智能在“办公AI Agent”上迈出的第一步,但并不是完全的一步,“这个软件会一直成长、生长,我们今天只是做了一个底座,这个底座上有一点点功能,但未来可能还有10个、100个功能,越来越全面。”

作为未来智能的CEO,马啸自认对产品早已了如指掌,但在全面迭代AI功能之后,这款耳机还是给了马啸很多意想不到的惊喜。

“原先我也不是特别在意数据的收集,这次耳机迭代升级之后,我经常戴一天,谈话、电话沟通都能直接录音,然后再把几款的常用办公软件打通数据,发现真的离不开了”,这时的AI耳机更像马啸的一个「小助手」——每到周末就可以收集上一周所有没决策的事项,然后排出优先级;把和部门leader之间沟通过,但没有完成的问题进行收集整理,提醒下周继续跟进;把公司整个经营数据、运营数据做整理和排序。

随着AI耳机越来越强大,外界更加关心未来智能的边界到底在哪里?

马啸曾在科大讯飞任职多年,早年敏锐地捕捉到职场人群“会议效率低下”的痛点后,他带领团队独立打造未来智能,长期死磕办公会议场景。经历5年的摸爬滚打,他依然坚持只做办公垂类,把办公场景做到极致。

这也是马啸的“进窄门、走宽路”理念,在办公这个“窄门”之内,持续深挖一公里,扩大办公类产品的品类。这次发布会上,马啸透露接下来还会有更多围绕着办公场景,打造的新形态产品陆续上市。

坦率而言,在办公领域,未来智能的AI耳机是先发者,已经占据了很多用户的心智,而新形态产品则将直面和飞书、钉钉等大厂的竞争,但马啸始终以平常心待之,他认为良性竞争不是零和博弈,而是大家共同培育一个新市场,加入的企业越多才能把这个“蛋糕”做得越大。

03 不争朝夕,做好基本盘

自ChatGPT诞生以来,AI潜移默化地进入人们的生活,资本热度更是一浪高过一浪。作为AI耳机的先行者,未来智能更是资本追捧的对象。

据了解,过去几年,未来智能陆续拿到了博裕创投、启明创投、蚂蚁集团等头部机构的投资。今年5月,公司又拿到了亿元级融资,“非洲手机之王”传音参投。

六轮融资之后,马啸依旧冷静,他融资的逻辑很简单——“够用就好”。

在马啸的规划中,投资方最好能和公司在业务层面产生协同。以蚂蚁为例,支付宝有很强的生活、出行、医疗等生态场景,未来智能和蚂蚁建立战略投资纽带,就有机会跟蚂蚁一起合作,由未来智能去做入口,支付宝的服务负责接入。

传音投资的逻辑同样绕不开合作。马啸提及,传音在非洲、印度等新兴市场建立了很强的分销体系,“我们会和传音有一些产品上的合作,也可以借助传音的营销渠道去接触更多用户,实现共赢。”

至于外界关注的上市话题,马啸觉得上市只是一个路径,“就是为了让公司有更多好的资金、更广阔资本的支持,去做出更多好产品。”

尽管资本一向着急,但马啸不希望自乱阵脚,“上市不是当下,时间表可能是未来几年。”

他不担心错过所谓的风口,“只要有好的思考、好的产品,顺应这个时代需求,在任何一个时代都有破局点。风口是在变,但是风口在的时候把用户服务得更好,当风口不在你这儿的时候,就做好你自己的基本盘,别把自己做到风口没了,掉地上就行。”

带着客户、服务至上的理念,未来智能逐渐走上世界的舞台。据马啸透露,未来智能目前用户数量已突破150万,遍布全球200多个国家和地区。

2025年,未来智能国内线下门店已超过3500家,核心商圈门店达2500家,出货总量突破25万件,单店产出同比增加56%。到了2026年一季度,海外市场全渠道合计增长3.8倍,活跃用户同比增长11.5倍,订阅用户同比增长高达23.8倍。这些数字说明,用户不仅愿意购买,更愿意持续使用和付费。

马啸从不过度包装自己和产品,他直率地讲述过去的迷茫,市场站位的优劣势。抬头看星空,方知自身之秒小;低头看脚下,才能一步一个脚印。在采访中,他频繁地提及“不着急”“平常心”“做好当下”,绝不冒进。

“窄门”既入,马啸便想日复一日把产品磨透,把AI藏进耳机、把用户装在心里。风总会停,但自己亲自踏出的路永不会消失。

注:文/王亚静,文章来源:创业最前线(公众号ID:chuangyezuiqianxian),本文为作者独立观点,不代表亿邦动力立场。

文章来源:创业最前线

广告
微信
朋友圈

这么好看,分享一下?

朋友圈 分享

APP内打开

+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