On April 1, the China National Intellectual Property Administration (CNIPA) issued a risk alert, stating that AI agents such as "OpenClaw" have been found to have vulnerable default security configurations. Using such agents to draft patent applications may trigger multiple risks. What are the specific security risks? How can they be prevented? Let's take a look at this quick Q&A.

1. Can using AI agents to draft patent applications lead to information leakage?

A: Agents like OpenClaw have risks such as excessive permissions, security vulnerabilities, and plugin poisoning. Using them to draft application documents can easily lead to the leakage of core information such as technical disclosure documents. Once leaked, the technical solution of the patent application may lose its novelty and become unpatentable, or worse, be preemptively filed by others, causing significant losses to the applicant. The agency involved may also bear liability for breach of contract and compensation.

2. Can a patent application with "AI hallucinations" still obtain protection?

A: Patent applications with "AI hallucinations" suffer from issues such as logical contradictions in content and unclear descriptions of technical features, making them unable to obtain protection.

3. What penalties apply for violating the principle of good faith?

A: Forming patent applications by generating, fabricating, or piecing together content through AI agents constitutes an act of bad-faith patent filing that violates the principle of good faith. Once a certain threshold is reached, applicants may face administrative penalties such as warnings and fines. Agencies and patent attorneys involved may face administrative penalties including revocation of their practice licenses and cancellation of their qualification certificates. In serious cases, they will be included in the list of serious violators of good faith.

4. How to prevent these risks?

A: For applicants: It is essential to increase risk awareness, carefully select compliant patent agency services, and proactively inquire whether the agency uses AI agents to draft application documents. If an applicant discovers that an agency has used such tools without authorization, resulting in information leakage or constituting an act of bad-faith filing, they may file complaints and reports in accordance with the law and demand compensation from the agency.

For patent agencies and patent attorneys: It is imperative to remain highly vigilant regarding the risks of using AI agents, refrain from using AI agents to engage in bad-faith patent filing practices, and effectively protect the legitimate rights and interests of their clients.

Source: Economic Daily News Client。