Recently, it has been exposed that agent tools including OpenClaw (nickname: Xiaolongxia; formerly known as Clawdbot and Moltbot) have vulnerable default security configurations, which may lead to severe security risks. Meanwhile, drafting patent application documents with such agents may trigger multiple potential hazards. In accordance with the risk notice issued by the China National Intellectual Property Administration （CNIPA）on April 1, 2026, using OpenClaw and similar agents to prepare patent application documents involves relevant risks. The detailed notice is as follows:

1.Risk of Technical Information Leakage: Agents such as OpenClaw suffer from excessive access permissions, security vulnerabilities, plug-in poisoning and other hidden dangers. Drafting application documents via these tools may easily result in the leakage of core confidential information including technical disclosure documents. Once leaked, the technical solutions of patent applications may lose novelty and thus fail to obtain authorization. In addition, third parties may file patent applications in advance, causing substantial losses to applicants. Patent agencies will also be liable for compensation for breach of contract.

2.Risk of Substantive Defects: When drafting application documents with such agents, AI hallucinations may occur, leading to logical contradictions in content, ambiguous description of technical features and other flaws in application documents, which will prevent the patents from being protected legally.

3.Risk of Dishonest Application: Filing patent applications with content fabricated out of thin air, randomly compiled or pieced together by intelligent agents constitutes dishonest patent application acts in violation of the principle of good faith. If such acts reach a certain frequency or quantity, applicants will face administrative penalties including warnings and fines. Patent agencies and patent attorneys may have their practicing licenses revoked and qualification certificates cancelled. Those with serious circumstances will be included in the list of serious dishonest entities for illegal acts.

To prevent the above risks and protect the legitimate rights and interests of all parties, further reminders are provided below:

1.For Applicants: Enhance awareness of risk prevention, carefully select compliant patent agency services, and actively confirm whether the agency uses intelligent agents to draft application documents. If an agency arbitrarily uses relevant tools and causes information leakage or conducts dishonest application acts, applicants may file complaints and reports in accordance with the law and claim compensation for losses from the agency.

2.For Patent Agencies and Attorneys: Stay highly alert to the risks of using intelligent agents, refrain from using such tools to conduct dishonest patent application practices, and effectively safeguard the legitimate rights and interests of clients.