Summary of Key Points
This article focuses on the "Louis Vuitton vs. Jasmine Milk White" trademark infringement case and, in conjunction with two lectures given by the author to legal researchers and lawyers, explores the capabilities and limitations of AI in legal work. It begins by clarifying common misconceptions in public opinion regarding the case (such as the supposed cross-category protection of well-known trademarks and a compensation amount of 4 billion yuan, which are both rumors). The article then explains the nature of AI using the concept of "probabilistic tools" (as opposed to deterministic tools like calculators, which produce definite results based on rules). It highlights that while AI can assist in legal tasks, it cannot replace humans in making value judgments, identifying flawed reasoning, or taking responsibility. The author emphasizes that in the age of AI, legal professionals must maintain essential skills such as independent thinking, cautious judgment, and a willingness to take responsibility.
Detailed Analysis
1. The Truth Behind the Case: Rumors vs. Facts
Many people discuss this case without having read the official judgment, leading to widespread misinformation. Here are some key misconceptions:
- Rumor 1: LV is using "cross-category protection of a well-known trademark" against Jasmine Milk White → Fact: Jasmine Milk White itself sold related products (handbags, earrings, scarves, etc.) within the categories registered by LV (e.g., handbags fall under LV's Class 18 trademark), so it's not a cross-category infringement.
- Rumor 2: The compensation amount is 4 billion yuan × 35% = 1.4 billion yuan → Fact: LV demanded 50 million yuan, and the court awarded only 10 million yuan (this includes franchise fees and brand management fees, with a discount for the contribution of the trademark) + an additional 300,000 yuan for legal costs; the figure of 4 billion yuan came from the media.
- Rumor 3: The court ordered Jasmine Milk White to apologize on all platforms and destroy its inventory → Fact: There was no requirement to destroy inventory; simply stopping the infringement was sufficient. Apologies were required only on the official website and five social media accounts, not across all platforms.
- Rumor 4: Jasmine Milk White only used the argument of "traditional patterns" in defense → Fact: The defendant's lawyers presented six lines of defense (public domain, non-trademark use, lack of similarity in services, no confusion, legal source, etc.), but the court refuted each one; they did not fail to make an effort.
In short, public opinion is often driven by emotions, while judgments are based on evidence and law. Don't let rumors influence your understanding of the facts.
2. AI Is Not a "Calculator," but a "Probabilistic Tool"
To understand what AI can do, it's important to understand its nature:
- Deterministic Tools: Tools like calculators always give definite results (e.g., 1 + 2 = 3). Legal databases provide accurate legal provisions and precedents when searched for relevant terms.
- Probabilistic Tools (AI): For the same issue, AI may generate multiple plausible outcomes. For example, when asked if the LV case was an infringement, AI could produce both supportive and opposing arguments. Its goal is to provide reasonable results, not absolute certainty—this means AI might create fictional case numbers or legal provisions that seem real but do not exist in reality.
This distinction is crucial: while AI offers opportunities (e.g., broadening perspectives), it also poses risks (e.g., the potential for creating false information).
3. How AI Can Assist Legal Professionals
In the LV case, AI could help with the following tasks:
- Data Organization: A 50-page judgment can be quickly structured by AI into a clear overview of seven trademarks, multiple batches of evidence, and timelines, saving time.
- Legal Research: AI can search for relevant Chinese trademark laws as well as similar cases in Europe, the U.S., and Japan, highlighting similarities and differences.
- Debate Simulation: AI can simulate roles such as plaintiff's lawyer, defendant's lawyer, judge, and scholar to highlight potential weaknesses in your arguments.
- Contradictory Analysis: AI can challenge your assumptions by translating intuitive notions into legal questions and generating both supportive and opposing arguments.
- Reference Verification: AI can list all cited legal provisions and precedents for you to verify their accuracy; however, it may also provide incorrect references, which you need to verify manually.
- Consumer Behavior Simulation: AI can create profiles of potential consumers to assess whether they might mistakenly associate Jasmine Milk White with LV products, but this is just for reference.
These are auxiliary tasks; AI does not replace human judgment.
4. Three Core Tasks That AI Cannot Perform
Despite its capabilities, AI faces three limitations:
- Value Judgment: Questions like "Where should the boundaries of trademark protection be drawn? Does a consumer's association with LV constitute infringement?" have no standard answers and depend on your priorities (e.g., protecting brands vs. promoting free competition). AI can provide arguments for both sides, but the final decision rests with humans.
- Identifying Flawed Reasoning: AI may create plausible arguments based on user preferences, but these may lack critical thinking (e.g., opposite legal opinions for the same case). You need to question the assumptions and potential exceptions.
- Taking Responsibility: You, as the author of the report or legal opinion, must bear the responsibility for any errors; AI cannot represent you in court or take blame.
5. What Legal Professionals Need to Do in the Age of AI
AI is not a panacea; it requires proper use:
- Evaluate AI Outputs: Don't accept AI results without questioning their basis. Treat them as drafts, not final conclusions.
- Maintain Human Decision-Making: Make critical decisions (e.g., case topics, strategies) independently. If you cannot do so without AI, you may be overly dependent on it.
- Interdisciplinary Collaboration: Understand the basics of AI to use it effectively (e.g., knowing how it might create false information).
In conclusion, while AI improves efficiency, it does not replace human judgment and wisdom. In the LV case, the issue was largely due to a lack of attention to the legal details (no one read the judgment). As with AI, the key is to maintain essential legal skills—reading the original documents, verifying evidence, and making independent judgments.
This article serves as a reminder that while AI is a valuable tool, fundamental legal competencies remain crucial. Just as in the LV case, the root of misunderstandings was the failure to engage with the legal facts.