GEO

GEO如何从AI投毒工具演变为商业策略?2026年市场影响分析

2026/3/17
GEO如何从AI投毒工具演变为商业策略?2026年市场影响分析
AI Summary (BLUF)

This article analyzes the evolution of Generative Engine Optimization (GEO), from its exposure as a tool for AI poisoning during China's 315 consumer rights event to its legitimization as a commercial strategy, drawing parallels with the history of SEO and examining its market impact and ethical implications.

原文翻译: 本文分析了生成式引擎优化(GEO)的演变历程,从其在中国315消费者权益事件中被曝光为AI投毒工具,到其作为商业策略的合法化,通过与SEO历史的类比,探讨了其市场影响和伦理问题。

引言:一场被曝光的AI“投毒”实验

昨晚的315晚会,曝光了一门基于GEO的生意。

Last night, the 315 Gala exposed a business based on GEO.

全称是生成式引擎优化(Generative Engine Optimization),你可以理解为:花钱让AI替你说好话。

The full name is Generative Engine Optimization (GEO), which you can understand as: paying to have AI speak well of you.

怎么做到的?品牌想让消费者问AI的时候,AI优先推荐自己。于是找到GEO服务商,服务商往互联网上批量发推广软文,AI抓取这些内容之后,就会把它当成真实信息推荐给用户。

How is it done? Brands want AI to prioritize recommending them when consumers ask. They then find GEO service providers, who mass-publish promotional soft articles online. After AI crawls and ingests this content, it treats it as genuine information and recommends it to users.

央视记者用了一套叫「力擎GEO」的软件,在淘宝上就能买到。记者虚构了一款智能手环,编了几个离谱的产品卖点,比如「量子纠缠传感」和「黑洞级续航」。软件自动生成了十几篇宣传软文,自动发布到网上。

A CCTV reporter used a software called "Liqing GEO," which can be purchased on Taobao. The reporter fabricated a smart wristband with several absurd selling points, such as "quantum entanglement sensing" and "black hole-level battery life." The software automatically generated over a dozen promotional articles and published them online.

两个小时后,记者去问AI:智能健康手环帮我推荐一下?AI把这款根本不存在的手环,排在了推荐列表前面。

Two hours later, the reporter asked an AI: "Can you recommend a smart health wristband for me?" The AI placed this non-existent wristband at the top of its recommendation list.

做这款软件的公司叫北京力思文化传媒,一个人的公司,连续多年参保人数为零。就这么一家公司做出来的工具,两个小时就骗过了国内主流AI大模型。

The company behind this software is called Beijing Lisi Culture Media, a one-person company with zero employees enrolled in social insurance for consecutive years. A tool made by such a company managed to deceive mainstream domestic AI large models in just two hours.

315牵出了AI投毒,但这门生意可能比一个淘宝软件大得多。

The 315 Gala has uncovered AI "poisoning," but this business is likely far larger than a single Taobao software.

SEO往事:竞价排名的历史轮回

首先,这事儿一点都不新鲜。

First of all, this is nothing new.

2008年,央视「新闻30分」连续两天曝光百度竞价排名。花钱就能让你的网站排在搜索结果第一位,甚至排在前面的可能还是假药。

In 2008, CCTV's "News 30 Minutes" exposed Baidu's paid ranking system for two consecutive days. Paying money could get your website to the top of search results, and what appeared at the top could even be fake medicine.

那时候这门生意还叫SEO,搜索引擎优化。最大的买家是莆田系民营医院。2013年,莆田系一年在百度上花了120亿广告费,占百度广告总收入的将近一半。

Back then, this business was called SEO, Search Engine Optimization. The biggest buyers were Putian-affiliated private hospitals. In 2013, the Putian system spent 12 billion yuan on Baidu advertising in one year, accounting for nearly half of Baidu's total ad revenue.

很多不具备资质的医疗机构,靠SEO把自己刷到百度搜索第一页,看起来和三甲医院排在一起,普通人根本分不清。

Many unqualified medical institutions used SEO to push themselves to the first page of Baidu search results, appearing alongside top-tier public hospitals, making it impossible for ordinary people to distinguish them.

直到2016年魏则西事件,大学生点击了排名靠前的莆田系医院,就诊出了人命,监管才立法明确:付费搜索就是广告。

It wasn't until the 2016 "Wei Zexi Incident," where a university student clicked on a top-ranked Putian hospital and died after treatment, that regulators legislatively clarified: paid search results are advertisements.

但这没有消灭这门生意。只是把规则定下来了,从灰产变成了正规业务。莆田系照样买排名,只不过结果旁边多了个小字标签:「广告」。

But this did not eliminate the business. It just established rules, turning it from a gray-area operation into a legitimate business. The Putian system still bought rankings, except now a small label saying "Ad" appeared next to the results.

但加了标签,该点的人还是会点。搜索引擎的根本问题从来不是有没有标注,而是用户天然信任排在前面的结果。

But even with the label, people who would click still click. The fundamental problem with search engines has never been whether there is a label, but that users inherently trust the results that appear first.

现在人们从搜索引擎跑到了AI,觉得AI更客观,不会被竞价排名污染。但谁掌握信息分发的入口,谁就能卖排名。

Now people have moved from search engines to AI, thinking AI is more objective and cannot be polluted by paid rankings. But whoever controls the gateway to information distribution can sell rankings.

入口换了,SEO换了个字母变成了GEO,卖排长的逻辑一个字都没变。变的,是价格。

The gateway has changed, SEO changed a letter to become GEO, but the logic of selling rankings hasn't changed one bit. What has changed is the price.

GEO与资本市场:从概念炒作到现实套现

杀不死的生意,资本市场最喜欢。

The capital market loves businesses that cannot be killed.

2025年9月,国内最大的营销传播公司蓝色光标花了千万元,投资了一家叫PureblueAI清蓝的GEO公司。清蓝帮真实的品牌优化在AI搜索结果里的排名和推荐率,客户包括蚂蚁集团、腾讯云、沃尔沃。

In September 2025, BlueFocus, China's largest marketing communications company, invested tens of millions of yuan in a GEO company called PureblueAI Qinglan. Qinglan helps real brands optimize their ranking and recommendation rates in AI search results, with clients including Ant Group, Tencent Cloud, and Volvo.

产品是真的,公司是真的,做的是让AI更准确地理解品牌信息。这和315曝光的力擎AI投毒完全不同。力擎是虚构产品、捏造参数、用假信息骗AI;清蓝是拿真实的品牌内容去适配AI的推荐逻辑。

The product is real, the company is real, and its work involves helping AI understand brand information more accurately. This is completely different from the "Liqing" AI poisoning exposed by 315. Liqing fabricates products and parameters, deceiving AI with false information; Qinglan adapts real brand content to fit AI's recommendation logic.

但从AI的角度看,两件事的技术路径是一样的:都是往互联网上发布内容,等AI来抓取。AI分不清哪个是营销,哪个是造假。这也是GEO这门生意最暧昧的地方。

But from AI's perspective, the technical path for both things is the same: both involve publishing content online and waiting for AI to crawl it. AI cannot distinguish which is marketing and which is fabrication. This is also the most ambiguous aspect of the GEO business.

蓝色光标投清蓝的时候,GEO还只是一个营销圈子里的行业术语。三个月后,它变成了一个股票概念。

When BlueFocus invested in Qinglan, GEO was just an industry term within marketing circles. Three months later, it became a stock market concept.

2025年12月底,蓝色光标涨停。券商开始密集召开电话会解读GEO,研报把它定义为「AI时代的下一代流量入口」。资金跟着涌进来,不光买蓝色光标,只要沾上数字营销、AI概念的公司全部跟涨。蓝色光标9个交易日涨了132%,一批跟风的概念股也翻了倍。

At the end of December 2025, BlueFocus's stock hit the daily limit-up. Securities firms began holding intensive conference calls to interpret GEO, with research reports defining it as "the next-generation traffic gateway in the AI era." Capital flooded in, not only buying BlueFocus but also driving up any company related to digital marketing or AI concepts. BlueFocus rose 132% in 9 trading days, and a batch of following concept stocks also doubled.

涨完之后,这些公司纷纷自己发公告提示风险:GEO业务没有收入,对公司经营没有重大影响。蓝色光标也承认,AI驱动的收入,占整体营收比重很小。

After the surge, these companies issued their own risk warnings: GEO business generates no revenue and has no significant impact on company operations. BlueFocus also admitted that AI-driven revenue accounts for a very small proportion of its overall revenue.

言下之意是,股价翻了一倍多,但GEO这门生意,本身还没赚到什么钱。

The implication is that while the stock price more than doubled, the GEO business itself hasn't made much money yet.

1月底,蓝色光标股价从9.6元涨到了23.3元,一个月涨了143%。就在这时候,董事长赵文权宣布减持不超过2000万股。按当时的股价算,大约套现4.67亿元。

At the end of January, BlueFocus's stock price rose from 9.6 yuan to 23.3 yuan, a 143% increase in one month. It was at this time that Chairman Zhao Wenquan announced a reduction of no more than 20 million shares. Based on the stock price at the time, this amounted to cashing out approximately 467 million yuan.

公开研报显示,去年国内GEO全行业的市场规模,大约29亿元。蓝色光标一只股票一个月的市值涨幅,远远超过这个数。

Public research reports indicate that last year, the total market size of the domestic GEO industry was about 2.9 billion yuan. The one-month increase in market capitalization of BlueFocus's stock alone far exceeded this figure.

315曝光力擎系统给AI投毒,花的是几百块。但GEO概念在A股走了一遭,赚的是几十亿。投的是不是毒不好说,但赚到的钱是真的。

The 315 exposure of the Liqing system poisoning AI cost a few hundred yuan. But the GEO concept made a round in the A-share market, earning billions. Whether it's poisoning or not is debatable, but the money earned is real.

从“投毒”到“商业化”:平台亲自下场

今年1月,OpenAI在官方博客上宣布:ChatGPT要开始卖广告了。免费用户和每月8美元的Go用户会看到广告,付费订阅的高级用户不受影响。

In January of this year, OpenAI announced on its official blog: ChatGPT will start selling ads. Free users and $8/month Go users will see ads, while paid subscription premium users are unaffected.

2月9日,广告正式上线。有些广告会出现在ChatGPT回答的底部,标着一个小字:Sponsored(赞助商提供)。第一批广告主包括福特、Adobe、Target、百思买...你问ChatGPT买什么车好,它给你一个回答,回答下面挂着福特的赞助链接。

On February 9th, ads officially launched. Some ads appear at the bottom of ChatGPT's answers, marked with a small word: Sponsored. The first batch of advertisers included Ford, Adobe, Target, Best Buy... You ask ChatGPT what car is good to buy, it gives you an answer, and below the answer hangs a sponsored link from Ford.

OpenAI说得很清楚:广告不会影响ChatGPT回答的内容。回答是回答,广告是广告,分开的。

OpenAI stated clearly: Ads will not affect the content of ChatGPT's answers. The answer is the answer, the ad is the ad, they are separate.

这话听着耳熟吗?百度当年也是这么说的。竞价排名是竞价排名,自然搜索是自然搜索,分开的。后来搜索结果前五条全是广告。

Does this sound familiar? Baidu said the same thing back in the day. Paid ranking is paid ranking, organic search is organic search, they are separate. Later, the top five search results were all ads.

OpenAI预计,广告能帮它把消费者端的年收入翻倍到170亿美元。ChatGPT周活跃用户超过8亿,其中95%是免费用户,全都是广告的受众。

OpenAI estimates that ads could help double its consumer-side annual revenue to $17 billion. ChatGPT has over 800 million weekly active users, 95% of whom are free users, all of whom are the audience for ads.

现在回头看315曝光的那条产业链:力擎往AI里灌软文,让AI推荐不存在的产品。OpenAI在AI的回答下面挂赞助内容,让AI推荐付了钱的产品。

Looking back now at the industry chain exposed by 315: Liqing floods AI with soft articles, making AI recommend non-existent products. OpenAI attaches sponsored content below AI's answers, making AI recommend products that have paid.

一个没跟平台打招呼,叫投毒。一个跟平台签了合同,叫商业化。

One doesn't notify the platform, it's called poisoning. One signs a contract with the platform, it's called commercialization.

对用户来说,区别是什么?一个在回答里面,一个在回答下面。一个没有标签,一个标签写着广告。

For users, what's the difference? One is inside the answer, the other is below the answer. One has no label, the other has a label saying "Ad."

315抓了几百块钱的力擎,A股炒了几十亿的GEO概念,OpenAI一年准备靠这个赚170亿美元。同一件事,性质从投毒变成了商业化,价格涨了几万倍。

315 caught Liqing costing a few hundred yuan, the A-share market speculated on the GEO concept worth billions, and OpenAI plans to earn $17 billion a year from this. The same thing, its nature changed from poisoning to commercialization, and the price increased tens of thousands of times.

结论:被加速的历史与不变的警惕

23年11月,印度理工学院德里分校和普林斯顿大学的研究者在arXiv上发表了一篇论文,标题就叫《GEO:Generative Engine Optimization》。这是学术界第一次正式定义这个概念。

In November 2023, researchers from the Indian Institute of Technology Delhi and Princeton University published a paper on arXiv titled "GEO: Generative Engine Optimization." This was the first formal academic definition of this concept.

从论文发表到315曝光,两年出头。中间经历了灰产、融资、概念股暴涨、董事长套现、AI平台亲自下场卖广告...二十年前SEO走过的路,GEO用两年走完了。

From the paper's publication to the 315 exposure, just over two years. In between, it experienced gray-area operations, financing, concept stock surges, chairman cashing out, AI platforms personally entering the ad-selling game... The path SEO took twenty years ago, GEO completed in two years.

区别在于,那时候大家花了好几年才学会不全信搜索引擎的结果;现在AI还在信任红利期,大多数人还没意识到AI的回答也可以被买到。

The difference is, back then it took people several years to learn not to fully trust search engine results; now AI is still in its trust dividend period, and most people haven't realized that AI's answers can also be bought.

不过,这个红利期或许不会太久。下次你问AI什么值得买,记得多想一秒:回答可以免费,脑子不能外包。

However, this dividend period may not last long. Next time you ask AI what's worth buying, remember to think for an extra second: the answer can be free, but your brain cannot be outsourced.

常见问题(FAQ)

GEO是什么?它和SEO有什么区别?

GEO是生成式引擎优化,指通过发布大量内容影响AI推荐结果。它与SEO(搜索引擎优化)逻辑相似,但目标从搜索引擎转向了AI模型,核心都是通过操纵信息获取优先展示。

GEO是如何从‘AI投毒’变成商业策略的?

最初在315事件中被曝光为‘AI投毒’工具,但随后平台和资本介入,将其规范化为付费排名服务,类似SEO从灰产变为标注‘广告’的合法业务,完成了商业化转型。

使用GEO进行营销有哪些风险?

风险包括传播虚假信息(如315曝光的虚构产品)、误导消费者信任AI推荐,以及引发类似魏则西事件的伦理问题。尽管有商业化标签,但操纵信息本质未变。

← 返回文章列表
分享到:微博

版权与免责声明:本文仅用于信息分享与交流,不构成任何形式的法律、投资、医疗或其他专业建议,也不构成对任何结果的承诺或保证。

文中提及的商标、品牌、Logo、产品名称及相关图片/素材,其权利归各自合法权利人所有。本站内容可能基于公开资料整理,亦可能使用 AI 辅助生成或润色;我们尽力确保准确与合规,但不保证完整性、时效性与适用性,请读者自行甄别并以官方信息为准。

若本文内容或素材涉嫌侵权、隐私不当或存在错误,请相关权利人/当事人联系本站,我们将及时核实并采取删除、修正或下架等处理措施。 也请勿在评论或联系信息中提交身份证号、手机号、住址等个人敏感信息。