广告
加载中

大厂AI 大战618

定焦One团队 2026-06-16 19:06
定焦One团队 2026/06/16 19:06

邦小白快读

EN
全文速览

本文梳理了今年618AI参与电商大促的全流程情况,核心干货如下:

1. 当前AI已经渗透到购物全环节:买前可通过AI助手完成选品、比价、领券、下单;交易中有24小时不打烊的AI数字人主播、秒回的AI客服,AI还能帮商家优化投放;买后AI优化库存调度和配送路线,大促也能更快收货。

2. 目前AI存在明显短板:仅能解决尺码、库存这类标准化问题,遇到优惠、保价、投诉等复杂问题,容易答非所问,还会阻断转人工通道;多数消费者不习惯AI直接给出购买决策,仅认可AI做比价、算优惠这类工具性工作。

3. 体验提示:遇到AI卡壳可多尝试发送转人工指令,复杂问题优先对接真人,AI导购可按需尝鲜不用盲从。

今年618各大平台集体押注AI,为品牌经营带来了新机会与新风险,核心干货如下:

1. 消费趋势层面:目前65%-70%的消费者接受AI做整理信息、比价、识别价格变动这类工具性服务,仅37%-48%接受AI直接给购买建议;消费者核心期待是AI帮商家降本让利、优化供应链减少缺货砍单。

2. 营销与运营层面:AI数字人直播成本低,24小时在线,京东免费开放使用,淘宝还给出流量倾斜扶持,但按规定数字人直播间必须标注AI生成,抖音等平台暂不支持外接数字人,违规会被限流甚至封禁。AI客服可降低七成人力成本,但仅能处理售前标准化问题,售后仍需人工介入。

3. 风险提示:AI回答错误引发的纠纷最终由品牌承担,使用平台AI工具需要交出用户数据,后续迁移成本较高。

今年618的AI布局,给电商卖家带来了明确的机会与风险提示,核心干货如下:

1. 机会层面:当前各大平台都开放了AI工具,京东免费开放数字人直播,抖音飞鸽智能客服限时免费,官方数据称可节省约70%的人力成本;AI还能帮中小卖家一键生成短视频和商品图文,降低运营门槛,AI可优化供应链库存周转效率30%到40%,降低运输成本,还能提供智能投放、需求洞察服务,帮助提升经营效率。

2. 风险提示:目前抖音、快手、小红书暂不支持外接数字人直播,未按要求标注AI生成会被限流封禁;AI仅能处理标准化售前问题,售后复杂问题仍需人工对接,AI回答出错的责任由卖家承担;深度使用平台AI工具后,迁移成本较高,卖家需要交出核心经营数据。

3. 长期机会:平台正在争夺AI消费决策入口,提前布局AI应用的卖家更容易获得流量倾斜。

今年618AI在电商全链路的落地,给工厂带来了数字化转型和业务增长的相关启示,核心干货如下:

1. 产品生产设计层面:当前AI已经可以打通消费需求洞察、内容生产到投放优化的全链路,工厂可以借助平台AI工具快速捕捉终端消费需求,反向指导产品研发和设计,让产品更贴合市场需求,减少盲目生产带来的库存压力。

2. 电商转型机会:AI降低了工厂做线上运营的门槛,AI数字人直播最低每小时仅12元,可实现24小时不打烊带货,成本远低于真人主播,一个数字人账号可通用多平台,适合工厂搭建自有线上销售渠道,降低线上运营的人力成本。

3. 供应链数字化启示:AI已经广泛应用在库存调度、运输路径规划环节,可提升库存周转效率30%-40%,降低干线运输成本,工厂推进数字化供应链改造,引入AI做供应链管理,可大幅提升大促等高峰时段的履约效率,降低整体运营成本。

今年618AI在电商大促的落地实践,给AI电商服务商指明了行业发展方向,核心干货如下:

1. 行业发展趋势:当前传统大促价格战的边际效果持续下降,大促模式陷入先涨后降的恶性循环,平台和商家都需要新的增长动能,AI是电商行业公认的核心增长点,行业判断2026年AI会成为电商的基建配套设施,未来两到三年就会迎来落地拐点,市场需求空间广阔。

2. 客户核心痛点:商家当前核心需求是降本,但现有AI产品仅能覆盖售前咨询、直播引流这类标准化场景,售后复杂问题的处理能力不足,同时现有产品出错率偏高,商家需要为AI的错误承担责任,这类痛点目前没有得到很好解决。

3. 产品方向:当前前端AI导购还处于探索阶段,消费者更认可AI的工具属性而非决策属性,服务商可围绕商家降本、合规、低出错的需求,打磨适配多平台的AI产品,抢先布局市场。

今年618各大平台的AI布局,总结出了平台发展AI的经验与方向,核心干货如下:

1. 商家与市场需求:当前商家对低成本降本增效工具的需求强烈,平台开放AI工具不仅可以满足商家需求,还可以赚取工具服务费,同时通过AI积累数据,能够加固平台自身的生态壁垒,形成强者恒强的循环。

2. 合规与现有可参考玩法:按照《直播电商监督管理办法》要求,AI数字人直播间必须在显著位置标注AI生成,否则需要做限流甚至封禁处理,平台需要做好规则管控规避合规风险。目前不同平台已经走出不同路线:阿里字节打通自有大模型和商城,把消费闭环锁在自有生态;京东全链路嵌入AI强化供应链优势;拼多多低调优化推荐算法,平台可根据自身定位选择适配路线。

3. 长期方向:AI争夺的核心是消费决策入口,平台需要借助大促逐步培养用户新的购物习惯,短期侧重降本提效,逐步探索前端智能导购场景,长期有望重构电商竞争格局。

今年618大厂AI参战大促,折射出电商产业的最新动向与待解问题,核心研究干货如下:

1. 产业新动向:传统大促价格战边际效益下降,大促模式陷入先涨后降的恶性循环,增长见顶,AI成为各大平台抢滩的新方向,今年AI已经从去年的锦上添花变成全链路渗透,行业判断短期AI侧重降本提效,长期会改变消费入口性质,未来两到三年会迎来落地拐点,2026年AI会成为电商基建,可能重新划分电商版图。

2. 新出现的产业问题:目前AI仅实现了原有流程的提效,没有改变电商现有商业模式;C端消费者要承担AI客服转人工难的隐性成本,AI降本并没有转化为商品价格下降让利于消费者;B端商家降本效果打折扣,还要承担AI出错的责任,交出数据后迁移成本极高;最终平台成为最大受益方,头部平台依靠数据优势进一步加固护城河,中小平台的竞争压力大幅提升。

3. 商业模式新变化:未来电商可能从搜索货架模式转向对话导购模式,平台从中立货架转变为消费决策分发方,竞争规则会发生根本性改变。

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 summarizes how AI participated across the entire shopping journey during this year's 618 mid-year shopping festival, with key takeaways as follows:

1. AI has now penetrated every stage of online shopping. Before purchasing, AI assistants can help users select products, compare prices, collect coupons and place orders. During the shopping process, 24/7 AI-powered digital human hosts and instant-response AI customer service serve shoppers, while AI also helps brands optimize ad placements. After purchase, AI optimizes inventory allocation and delivery routes, enabling faster delivery even during the peak promotion period.

2. AI still has clear limitations today: it can only handle standard queries such as size and inventory checks, but often gives irrelevant answers to complex questions about discounts, price protection, or complaints, and may even block users' access to human customer service. Most consumers are not accustomed to letting AI make direct purchase decisions, and only accept AI for tool-oriented tasks such as price comparison and discount calculation.

3. User tip: If AI gets stuck, try sending a request to connect to a human agent multiple times; prioritize human service for complex issues, and try AI shopping assistants on an as-needed basis instead of following their suggestions blindly.

This year's 618 saw all major e-commerce platforms bet heavily on AI, bringing new opportunities and new risks for brand operations, with key insights as follows:

1. On consumer trends: Currently, 65% to 70% of consumers accept AI for tool-based services such as information sorting, price comparison and price change tracking, while only 37% to 48% accept AI giving direct purchase suggestions. Consumers' core expectation is that AI will help brands cut costs to lower prices and optimize supply chains to reduce out-of-stock issues and order cancellations.

2. On marketing and operations: AI-powered digital human live streaming costs far less than human hosts and operates 24/7. JD.com offers free access to this tool, and Taobao even provides additional traffic support for digital human streams, but regulations require all AI-generated digital human live rooms to be clearly labeled. Platforms including Douyin do not allow third-party external digital human solutions, and violations will lead to traffic restrictions or even account bans. AI customer service can cut labor costs by 70%, but only handles standardized pre-sales queries, with after-sales issues still requiring human intervention.

3. Risk warning: Brands ultimately bear liability for disputes caused by incorrect AI responses. Using platform-provided AI tools requires sharing user data, which leads to high migration costs if brands want to switch systems later.

AI deployments during this year's 618 have delivered clear opportunities and risk warnings for e-commerce sellers, with key takeaways as follows:

1. Opportunities: All major platforms have now opened AI tools to sellers. JD offers free access to digital human live streaming, and Douyin's Feige intelligent customer service is available for free for a limited time. Official data shows these tools can cut labor costs by around 70%. AI also helps small and medium-sized sellers generate short videos and product copy and images in one click, lowering the barrier to entry for operations. It can improve supply chain inventory turnover efficiency by 30% to 40%, cut shipping costs, and provide services such as smart ad placement and demand insight to boost overall operational efficiency.

2. Risk warning: Currently, Douyin, Kuaishou and Xiaohongshu do not allow third-party external digital human live streaming; failing to label AI-generated content as required will lead to traffic restrictions or account bans. AI only handles standardized pre-sales queries, and complex after-sales issues still require human support. Sellers bear full liability for incorrect AI responses. After deep integration with platform AI tools, migration costs become very high, and sellers have to share their core operational data with platforms.

3. Long-term opportunity: Platforms are now competing to control AI-powered consumer decision-making entry points, so sellers that deploy AI applications early are more likely to receive favorable traffic allocation.

The end-to-end deployment of AI in e-commerce during this year's 618 offers key insights for factories on digital transformation and business growth, as follows:

1. For product development and design: AI now enables end-to-end connectivity from consumer demand insight, content creation to ad placement optimization. Factories can leverage platform AI tools to quickly capture terminal consumer demand, and use those insights to guide product R&D and design, making products more aligned with market needs and reducing inventory pressure caused by blind production.

2. Opportunities for e-commerce transformation: AI lowers the barrier to online operations for factories. AI digital human live streaming costs as little as 12 yuan per hour, enables 24/7 sales, and costs far less than human hosts. A single digital human account can be used across multiple platforms, making it ideal for factories to build their own online sales channels and cut online operational labor costs.

3. Insights for digital supply chain transformation: AI is already widely used for inventory scheduling and transportation route optimization, improving inventory turnover efficiency by 30% to 40% and reducing trunk transportation costs. Factories that advance digital supply chain transformation and adopt AI for supply chain management can significantly improve fulfillment efficiency during peak periods such as 618, and cut overall operational costs.

The real-world deployment of AI during this year's 618 shopping festival points to clear industry development directions for AI e-commerce service providers, with key takeaways as follows:

1. Industry development trends: The marginal effect of traditional promotion price wars continues to decline, and the current promotion model is trapped in a vicious cycle of raising prices before cutting them. Both platforms and brands need new growth drivers, and AI is widely recognized as the core growth engine for the e-commerce industry. Industry analysts forecast that AI will become a core infrastructure for e-commerce by 2026, and the industry will reach an inflection point for large-scale adoption within the next two to three years, creating vast market demand.

2. Core customer pain points: Brands' top priority right now is cost reduction, but existing AI products only cover standardized scenarios such as pre-sales consultation and live streaming traffic acquisition, and lack the capability to handle complex after-sales problems. Meanwhile, existing products still have high error rates, and brands have to take responsibility for AI mistakes. These pain points remain largely unaddressed today.

3. Product development direction: Front-end AI shopping guides are still in the exploration stage, as consumers recognize AI more as a tool than a decision-maker. Service providers should focus on developing multi-platform compatible AI products that meet brands' needs for cost reduction, compliance and low error rates, to capture an early foothold in the market.

AI deployments across major platforms during this year's 618 offer key lessons and development directions for e-commerce platforms looking to build out AI capabilities, as follows:

1. Merchant and market demand: Merchants have strong demand for low-cost tools to cut costs and boost efficiency. Opening AI tools to merchants not only meets this demand, but also allows platforms to earn tool service fees. Meanwhile, AI enables platforms to accumulate more user and operational data, which strengthens their own ecosystem barriers and creates a "winner-takes-all" cycle.

2. Compliance and proven operating models: According to the *Administrative Measures for the Supervision and Administration of Live Streaming E-commerce*, AI-generated digital human live rooms must be clearly labeled as AI-generated in a prominent position, otherwise platforms must implement traffic restrictions or even bans. Platforms need to implement strong rule management to avoid compliance risks. Today, different platforms have adopted distinct strategies: Alibaba and ByteDance have integrated their proprietary large language models with their own e-commerce marketplaces to lock the entire consumer journey within their ecosystems; JD has embedded AI across its entire operation to strengthen its supply chain advantage; Pinduoduo has quietly optimized its recommendation algorithm. Platforms can choose a path that fits their own positioning.

3. Long-term development direction: The core competition around AI is control of consumer decision-making entry points. Platforms need to gradually cultivate new user shopping habits through large promotion events, focusing on cost reduction and efficiency improvement in the short term while gradually exploring front-end intelligent shopping guide scenarios. In the long run, AI is expected to reshape the competitive landscape of the e-commerce industry.

The participation of major tech companies' AI in this year's 618 shopping festival reflects the latest trends and unresolved issues in the e-commerce industry, with key research insights as follows:

1. New industry trends: The marginal benefit of traditional promotion price wars is declining, and the promotion model is trapped in a vicious cycle of raising prices before cutting them, with growth topping out. AI has become the new frontier that all major platforms are racing to capture. This year, AI has shifted from being a "nice-to-have add-on" last year to penetrating the entire e-commerce journey. The industry consensus is that AI will focus on cost reduction and efficiency improvement in the short term, reshape the nature of consumer entry points in the long term, reach an inflection point for large-scale adoption within two to three years, become core e-commerce infrastructure by 2026, and potentially redraw the e-commerce map.

2. New emerging industry problems: Currently, AI only improves the efficiency of existing processes and has not changed the existing e-commerce business model. C-end consumers bear the hidden cost of difficulty connecting to human customer service after AI interaction, and cost savings from AI have not been passed on to consumers in the form of lower prices. For B-end merchants, the cost reduction effect is limited, and they still have to bear liability for AI errors, while sharing data with platforms leads to extremely high migration costs. Ultimately, platforms are the biggest beneficiaries: leading head platforms rely on their data advantages to further strengthen their moats, significantly increasing competitive pressure on smaller mid-sized platforms.

3. Potential changes to the business model: E-commerce may shift from the search-based shelf model to a conversational shopping guide model in the future. Platforms will transition from being neutral marketplaces to consumer decision distributors, which will fundamentally change the rules of competition.

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.

AI让生意更好做了吗?

定焦One(dingjiaoone)原创

作者 | 李梦冉

编辑 | 魏佳

往年618的头等大事,是背满减公式。“跨店每满300减50”,让消费者都临时变成奥数考生,社交平台上到处是“数学课”和“抄作业”帖,大家对着购物车反复计算,生怕买贵了。

今年618,有些不同。

从开启时间看,战线一如既往地长:抖音5月15日开跑,拼多多从5月21日持续到6月30日,天猫5月21日开卖,京东5月30日晚8点启动。玩法却明显简化,跨店满减、定金膨胀这些传统促销还在,但平台不再把它们当作传播重心,真正的C位,让给了AI。

千问和淘宝全面打通,打开千问APP,可以让AI帮你挑选淘宝商品;豆包APP打通抖音商城,内嵌的“买前问豆包”功能可以直接下单;京东的AI数字人主播在凌晨三点仍在讲解商品......

事实上,AI介入大促并不是第一次。去年双11,各大平台就已尝试用AI生成商品文案、做智能客服和直播数字人。只是当时更像“锦上添花”,而今年618,AI几乎出现在了每一个环节。

平台战报也在强调它的存在感。京东618前4小时AI数字人带货突破7000万元;淘宝升级AI店小蜜,上线“AI假图识别模型”,平均退款挽单成功率超过20%。

虽然这些数字在大促万亿级GMV中算不上主角,但在价格战打了好几年,边际效果下降的背景下,是一个新信号。只是,AI到底是“神器”,还是各平台请来撑场面的新道具?

01.AI加入618:直播不打烊,客服秒回,物流更快

要形容今年618中AI扮演的角色,“无处不在,但并非无所不能”或许最贴切。从消费者打开APP的第一秒,到商品出库的最后一公里,AI几乎渗透进了所有环节。按购物的流程,可以拆成买前、交易中、买后三个阶段来看。

买之前,入口端变化最明显。过去,购物的起点是搜索框:输入关键词,系统抓取商品列表,再靠自己筛选。今年618,多家平台完成了AI助手与购物入口的打通,起点从“搜索”转向“对话”。

淘宝与千问全面打通,用户在千问APP里表明需求,千问可以完成选品、比价、领券甚至下单;抖音则将豆包APP与商城打通,内嵌“买前问豆包”功能,用户问答后可直接跳转购买,全程不离开抖音生态。

不过,AI助手的使用目前仍以尝鲜用户为主,远未成为主流购物方式。

交易之中,商家的感受最直接,AI主要体现在直播、客服和经营决策三块。

直播是各家都在发力的环节之一。京东数字人“JoyStreamer”今年618期间开播数量同比增长6倍,超过7万商家在使用。这些数字人主播可以7×24小时在线,凌晨三点仍能讲解、促单、发券。淘宝、拼多多、唯品会也都开放了数字人直播接入,淘宝对数字人直播还推出激励政策,包括流量倾斜和技术扶持等。不过不同平台对数字人的态度和执行尺度不一,今年2月《直播电商监督管理办法》正式施行后,所有平台的数字人直播间都须在显著位置标注“AI生成”标识,否则将面临限流甚至封禁。

客服端也同样用上了AI。淘宝的“店小蜜”从自动回复升级为“导购型”客服,能主动推荐商品,还上线了“AI假图识别模型”拦截恶意退款,商家测试数据称平均退款挽单成功率超20%;京东AI客服“京小智”已服务超百万商家;抖音的飞鸽智能客服向商家限时免费开放,官方披露可节省约70%的人力成本。

此外,AI工具开始介入更深层的经营决策。如淘宝的“AI万相”通过多智能体协同,打通需求洞察、内容生产和投放优化;抖音的“千川·乘方”则提供智能投放服务。不少平台还开放了AIGC工具,帮中小商家一键生成短视频和商品图文。

以上这些,都是“看得见”的环节。买完之后,是消费者感知最弱的物流与供应链环节,但这恰恰是AI应用最扎实的一环。

京东物流今年重点推进的“超脑大模型”,覆盖超过1000个核心供应链场景,库存周转效率提升30%至40%,干线运输成本下降;淘宝、拼多多等电商平台甚至各大物流体系也利用AI做智能调度、动态分配库存。

当你下单后,AI会预测你所在地的库存需求,自动从最近的仓库调货,同时规划最优配送路线。这也是为什么在大促期间,有些消费者下单后几小时即可收到货,背后是供应预测链在默默支撑。

总的来说,AI确实在改变618的每一个环节,入口更聪明,直播不打烊,客服秒回,物流更快。但这些大多都停留在提效层面,原有的流程变得更快,却对商业模式没有改变。平台战报里的漂亮数字,和消费者、商家的真实体验之间,或许还有距离。

02.AI,让生意更好做了吗?

消费者端,AI客服是大家感受最明显也最有争议的地方。

社交平台上,关于“AI客服快把人逼疯了”的话题下挤满了吐槽。消费者夕夕的反映尤为激烈,她告诉「定焦One」,大促期间咨询量暴增,AI客服顶在最前面,能解决的问题却很有限。“尺码、库存、发货时间这些标准问题还能回答,一旦涉及优惠活动、保价、投诉,AI要么答非所问,要么套话循环。”她曾连续发了十几条“转人工”,都没能对接上真人客服。

至于AI导购、AI帮你选,她从未用过。“我是个目的性很弱的购买者,喜欢‘逛’的感觉。如果目的明确,我会直接搜索,买最划算的链接,不需要AI替我决定。”

这种心态不是个例。一份618前瞻调研显示,在提到对于AI赋能消费的具体期待中,前三名都集中在提效上——51.7%的人希望“AI帮商家提效,降低成本让利给我”,50.3%希望“AI优化供应链减少缺货砍单”,44.0%希望“AI提升售后客服效率”。

尤其是,消费者对“整理信息、比价、计算优惠、识别价格变化”等工具性功能接受比例在65%-70%,但对“AI直接给出购买建议或自动下单”则降至37%-48%。

看来,目前消费者愿意听AI建议的并不多。

商家这边更复杂:一边是服务商给出的漂亮数据,一边是一线运营的真实顾虑。

先看成本这笔账,AI确实把门槛拉低了。

一位AI数字人服务商称,目前他们AI数字人直播报价低至每小时12元,这意味着,24小时不间断直播每天不到300元,远低于一个真人主播的时薪。而且一个账号可以在京东、淘宝、拼多多等多平台通用。京东甚至向商家免费开放了数字人直播,但是功能与外接付费版本存在差距。

AI客服同理。电商平台官方AI客服基础功能通常免费或提供免费额度,但高阶功能、超量使用或企业级定制会收费。

某外接AI客服供应商给「定焦One」的报价单显示,入门版年费16800元,加上平台接入费(此费用由电商平台收取,不同平台价格不一),如淘天接入费360元/年/个,实际年成本约17160元。这显然远远低于一个全职客服的年薪,还能做到全天在线、随时响应。

但账面上的便宜,落到实际场景里就没那么理想了。

一方面,能用上的范围有限,且各平台在AI直播上的进展并不一致。上述AI数字人服务商告诉「定焦One」,电商平台里,目前京东、淘宝、天猫、拼多多、唯品会可以外接数字人直播,抖音、快手、小红书还不行。甚至有商家反映,在某些平台上数字人直播有被限流甚至封禁风险。

另一方面,在一些场景里,即便用上了AI,不代表不需要真人客服。上述AI客服供应商透露,售前咨询用AI,成本压缩效果明显,但对于售后问题,AI处理得没那么精准,需要人工接手。尤其大促期间咨询量激增,反而是AI最容易暴露短板的时候。

更让商家迟疑的,是责任问题。

一位零食类目电商商家对「定焦One」直言:AI抓取的数据都是根据你产品本身已有的竞争力,产品好就能被AI抓取,所以“AI降本作用不大,但增效是有的”。

对于电商平台的AI工具,她都有所了解,但真正信任的不多,“毕竟AI不需要承担后果。”如果AI客服回答错误引发纠纷,最后背锅的还是商家。

那么,AI省下的成本和效率,最终到底流向了谁?

商家确实降低了人力成本,但也交出了一部分数据和决策权。工具越顺手,迁移成本越高。

消费者则承担了一部分隐性成本。AI客服替商家省了钱,却让用户陷入“转人工”死循环;所谓的“假性降价”依然存在,AI省下的效率并没有变成更低的价格。

相比之下,平台受益更明显。除了可以赚取工具费,还能获得数据和生态壁垒。AI越智能,越需要数据喂养,数据越集中,AI越强大,这是一种强者恒强的循环,最终加固了平台的护城河。

03.平台押注AI:短期被高估,长期被看好

AI在今年618的实际效果还很一般,平台为什么还要集体押注?这要从两方面来看。

一边是大促的吸引力在下降,平台需要一个新故事。

一位资深电商运营告诉「定焦One」,他所负责的某类目在2025年618期间还有70%的增长,但今年和平时数据差不多,几乎没有明显增长。在他看来,有部分原因是大背景变化,“这是靠运营无法解决的。”

但更深层的原因在于大促模式本身,“平台喜欢大促,商家为了利润,消费者为了优惠,双方诉求很难同时满足,商家被夹在中间。”

他告诉「定焦One」,为了应付平台的大促活动同时保住利润,一些商家会“先涨后降”。假设这一年都参加平台的活动,有20%的优惠,那商家就把价格拉高,再通过活动回到日常价。尤其是,如今促销活动愈发密集、战线越拉越长,让消费者对价格越来越敏感,商家只好“假性降价”,这是一种恶性循环。

当补贴的边际效果越来越弱、大促模式本身陷入内耗,平台急需一个新的增长故事,AI就被推到了台前。

另一边,被推上台的AI,对平台的意义远不止“降本增效”,这从各家的布局就能看出来。

今年618,各平台对AI的投入力度和打法都不同,方向却一致。阿里和字节的思路相似,用自家大模型把消费决策的闭环锁在自家生态内;京东走的是全流程嵌入,强化履约效率和供应链成本优势;拼多多最低调,不追AI概念,默默做价格匹配和推荐算法的优化。

艾媒咨询首席分析师张毅向「定焦One」分析,平台布局AI的核心,都是在抢占消费决策的入口。“电商竞争由流量竞价变为用户心智的争夺,阿里、字节、京东都依托各自生态构筑壁垒。”他同时指出,这对没有生态依托的中小平台而言,是巨大压力。

入口之所以重要,因为AI正在改变入口的性质。搜索时代,用户主动检索、自己比价,平台更像一个中立的货架;而到了对话时代,AI替用户做筛选和推荐,平台的角色,从货架变成了导购,议价能力完全不同。谁掌握了那个被用户信赖的“AI助手”,谁就掌握住了消费决策的分发权。这也是阿里要把千问和淘宝打通、字节要把豆包和抖音商城打通的原因。它们抢的不是一次大促的GMV,是借着大促培养用户新的购物习惯。

张毅认为,2026年的618,AI已经是“基建配套设施”。眼下各平台侧重降本,只是短期的阶段性特征。“目前AI大多用于后端的运维,前端智能导购还在探索阶段。”

这也解释了为什么今年618消费者对AI的感知并不强烈,消费者能直接接触到的“前端”变化,还没有真正成熟。

短期来看,AI对GMV的直接拉动未必明显。它的价值更多体现在转化率提升、客服成本下降、投放效率优化、履约路径改善。但长期来看,这是一场关于“购买入口”的争夺,甚至可能重新划分电商版图。

“AI短期落地价值会被高估,但长期的潜力还有更大的提升空间。”张毅判断,拐点可能在未来两到三年出现,随着多模态智能体和供应链的落地,未来C端的AI一键购物、B端的AI产销改造、大商家出现分层发展,都是可以预见的趋势。

但至少今年618,它还在路上。

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

文章来源:定焦One

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