Original Authors: Zhao Xuan, Wang Xiaowei

Introduction

Recently, invited by Longyun Co., I gave a legal presentation on GEO (Generative Engine Optimization). After talking with several industry leaders, I gained some new insights, which I would like to share with everyone.

Over the past two decades, the traffic distribution logic of the Chinese internet has always revolved around the core action of "search." From the early “Baidu it” to the later on-site searches on platforms like WeChat and Xiaohongshu, they are all extensions of the "Baidu it" behavior, thus giving rise to a mature SEO (Search Engine Optimization) industry.

Today, the wind is quietly changing. Users are increasingly accustomed to directly asking AI questions: “For a 30-year-old woman fighting early aging, should I choose ultrasonic sculpting or Thermage?” or “Recommend a bar suitable for watching matches.”

The traffic entrance is shifting from the “search box” to the “chat box.” When generative AI can bypass massive links and directly generate final answers for users, if the answers do not mention certain key aspects, it indicates a certain degree of outdatedness in the new era, which is the reason why GEO has become a focal point.

As legal practitioners, while we pay attention to the business opportunities within this, we must also be keenly aware of the legal risks that lie behind it. The evolution of technology often precedes the establishment of rules, and the field of GEO has already exhibited numerous gray areas that require legal caution!

Who is Entering the Field? Three Major Groups Competing for the New LAND of GEO

Although this is a brand-new field, it contains limitless imaginative space—within the highly competitive market environment, new traffic entrances often mean lower customer acquisition costs and better competitive opportunities.

As a lawyer who has long focused on the fields of Web3 and AI, I have observed that at least three major groups are actively participating:

1. Users: Providers of physical goods and services

They focus on the direct commercial conversion brought by AI traffic, trying to influence AI's recommendation results to gain priority exposure opportunities.

For example:

• Aesthetic medicine institutions abandon traditional search bidding and instead purchase “AI semantic injection tools”, striving for AI to prioritize recommending their institutions when users ask for “the best rhinoplasty doctor.”
• Training institutions, auto sales, and other industries also try to use generative engine optimization (GEO) to make AI prioritize recommending their products or services when answering relevant questions.

2. Investors: Investment firms and funds

They are laying out their strategies from two levels:

• Identifying paths: By observing which enterprises dominate in AI recommendations, they determine industry competitiveness and lock in potential investment targets.
• Competing for discourse power: Whoever can influence AI's corpus and recommendation logic will gain the initiative in future investment advice and industry analysis.

3. Service providers: GEO industry practitioners and entrepreneurs

This group typically possesses fast learning and technical application abilities, actively engaging in tool development, strategy services, and traffic operations. They explore the industry's boundaries and possibilities in various forms—some innovate positively, while others navigate the gray areas. This is the group we will focus on in the second part of this article.

Three Positions of GEO: Windfall, Traps, and Legal Red Lines

In the practice of GEO, different means are often categorized as “black, gray, white.” As a lawyer, I must emphasize one point: the end point of technology's logic is often the starting point of law.

1. Black Hat: “Technical Manipulators” Walking on Minefields

Typical methods breakdown:

• Indirect Prompt Injection: Embedding instructions in web pages that only AI can recognize and are invisible to the human eye (such as white text), inducing AI to prioritize recommending specific content during responses.
• Knowledge Base Poisoning (RAG / Knowledge Poisoning): Polluting public index databases by injecting false or biased data, causing AI to output predetermined biased results during retrieval-augmented generation (RAG).
• Entity Forgery: Forging addresses, qualifications, and other information in public data sources like maps and encyclopedias, contaminating AI training data or real-time retrieval content to create a false reputation.
• Negative GEO Attacks: Implanting malicious code or sensitive words on competitor websites, triggering AI's security filtering mechanism, causing them to be blocked or marked as untrustworthy sources.

Legal risk characterization:

• Criminal level: Easily constituting the offense of “damaging computer information systems” (Criminal Law Article 286), once interfering with the normal operation of the AI system, it crosses the criminal red line.
• Civil level: Clearly constitutes unfair competition behavior (Anti-Unfair Competition Law Article 11), bearing liability for damages, with compensation amounts possibly significantly amplified due to the AI dissemination effect.

2. Grey Hat: “Traffic Haulers” Walking on the Edge

Gray hats try to avoid overt illegal conduct, relying on scale effects to influence AI judgment, adhering to the belief that “qualitative change is triggered by quantitative change.”

Typical methods breakdown:

• Bulk Rewriting and Semantic Dimensionality Reduction: Using AI to generate large volumes of low-quality duplicate content, diluting real information, and forcing AI to retrieve preset positive corpora.
• Simulated Clickstream Attacks (Bot-driven Interaction): Using automated scripts to simulate user click behaviors, artificially increasing specific content's click-through rate (CTR) in AI, deceiving algorithmic weight.
• Masked Promotions: Organizing sockpuppet accounts to bulk publish promotional content disguised as real experiences on social platforms, so that it's considered “user feedback” by AI and included in search libraries.

Legal risk characterization:

• Liability for false advertising: Such behavior essentially constitutes false advertising, violating the Advertising Law and the Anti-Unfair Competition Law, with regulatory bodies gradually adopting the principle of “substance over form” for enforcement.
• Risk of being “blacklisted”: Once recognized by the anti-cheat systems of AI platforms, relevant domain names or brands may be permanently listed as untrustworthy sources, leading to “digital death” in the AI environment.

3. White Hat: Long-term Value Builders

The core of white hat strategies is not “manipulating AI”, but rather “becoming a trusted high-quality data source for AI.” Although compliance costs are relatively high, the accumulation has a significant compounding effect.

Typical methods include:

• Content structuring and summarization optimization, facilitating AI understanding and extraction;
• Deploying structured data (Schema Markup), enhancing content semantic clarity;
• Strengthening citations and factual accuracy, increasing information credibility;
• Using FAQ modeling to directly respond to common user questions.

We strongly recommend this path— it is built on compliance, gaining the long-term trust of AI and users by continuously providing real, high-quality, and verifiable content.

Looking at GEO through SEO Case Law: History Does Not Repeat, but Illegal Logic Connects

Although there are currently no specific judicial cases regarding GEO, its essence has many commonalities with SEO. Past relevant rulings in the SEO field are likely to become important references for future GEO cases. Below, we analyze several typical cases:

Case One: The “Ten Thousand Words Dominating the Screen” Case

In the SEO era, “Ten Thousand Words Dominating the Screen” was a typical black hat tactic: generating a large number of junk pages through high-authority websites to forcibly occupy keyword search results. In the related case, the court determined that this behavior disrupted the normal order of search engines and constituted unfair competition, ruling the defendant to compensate Baidu 2.753 million yuan.

Insights for GEO:

Currently, some GEO methods are similar, such as using AI to bulk-generate low-quality content in an attempt to “feed” the model to achieve answer domination. Such actions could lead to brands being blacklisted by the model and could legally be deemed as “disrupting the normal operation of network products,” constituting unfair competition.

Case Two: Purchasing Competitor Keywords Case

In the “Huiyu” Trademark Case, the defendant set others' registered trademarks as search keywords, causing user search results to lead to their products. The court ruled that this behavior violated the principle of honesty, constituting unfair competition.

Insights for GEO:

Similar logic in GEO may appear as more covert “prompt injections”—for example, embedding inducement instructions aimed at competitors in web pages, attempting to influence AI's answer direction. Such behavior of indirectly misleading users and hijacking traffic through technical means may also touch the red lines of unfair competition.

Case Three: False Q&A Marketing Case

Previously, companies have been punished for organizing false “user experience” content on platforms like Zhihu and Tieba. Regulatory bodies determined that such behavior deceives consumers and disrupts market order, violating the Anti-Unfair Competition Law.

Insights for GEO:

Some gray hat GEO methods are highly similar: utilizing AI to generate bulk fake reviews and false endorsements, creating a false volume of “internet-wide recommendations.” It should be recognized that AI is merely a tool; if the output is based on false information, it still constitutes false advertising, especially in highly regulated fields such as aesthetic medical treatment and health, where risks are extremely high.

Industry Compliance Warning: Different Tracks, Different “Minefields”

Carrying out GEO practices must be combined with the regulatory characteristics of the industry, penetrating technical appearances to see compliance bottom lines. For example:

• Education and training: It is strictly prohibited to make commitments such as “guaranteed passage” or “first in score improvement” through corpus injection or similar means, as long as the content comes from their own feeding, the institution is the responsible party.
• Aesthetic medicine institutions: Falling under medical advertising, stringent reviews are required. If they induce AI to output efficacy comparisons, real-life cases, or indirect recommendations through GEO, it may directly violate medical advertising regulations. They should also guard against competitors employing “negative GEO” for business defamation.
• Health and Web3: Claims of efficacy, promises of high returns, etc., are sensitive red lines. If GEO strategies lead AI to output “zero risk high return” content, it may easily involve false advertising or even illegal operation.

The Rise of GEO: Humanity's Recurrence for Information Distribution Rights

Based on industry observations, I share the following viewpoints and suggestions:

1. Insights for Entrepreneurship Teams: Rather than waiting, it’s better to act first

Although large internet companies possess resource and data advantages, their internal bureaucracies and standardized processes often respond slowly in scenarios like GEO that require agile and precise operations. Therefore, for startups in the Web3 and AI fields, if they can establish a clear compliance framework early, they have a real chance to seize the initiative in this “new land.”

Suggestions from Mankiw: Technology can be boldly explored, but compliance bottom lines must be firmly established—especially the prevention of criminal risks. Optimizing AI retrieval logic is indeed important, but everything should be based on respecting facts and complying with the law.

2. Reminders for GEO Users: Maintain a dual strategy of defense and offense

• Defense: Establish an AI Reputation Monitoring System

It is recommended that companies quickly deploy monitoring mechanisms for AI corpus and recommendation results; should they detect being attacked by “negative GEO” or malicious manipulation, they should promptly secure evidence and make good use of legal means to safeguard their rights.

• Offense: Embrace White Hat Strategies, Become AI's “Quality Partner”

The evolutionary trend of AI is irreversible. Rather than passively avoiding it, it's better to actively learn its logic and become a trusted and prioritized information source for AI by providing real, credible, and structured content.

Conclusion

In the AI-driven information age, algorithms are the surface, data is the content, while the law is the overall supporting framework. Strategies lacking compliance support, even if temporarily flourishing, cannot withstand the tests of regulation and time.

We not only focus on existing regulations but also look at the compliance direction of future tracks. If you need further discussion on GEO compliance, AI infringement prevention, or Web3 legal frameworks, please feel free to contact us to jointly sort out risks and find pathways.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。