Generative AI in patent practice: the synergy between innovation and human expertise

To remain at the forefront of this technological revolution, we closely follow and investigate developments in AI applications on the market. However, we go a step further than merely adopting existing commercial software; we also actively develop our own generative AI tools in-house. By tailoring these applications specifically to the nuances of patent law, we create customized solutions. This enables us, for example, to better analyze relevant prior art, brainstorm technical approaches, or more efficiently process extensive technical documentation.

Despite our positive attitude toward these innovations, we guard against blind dependence on algorithms. In our practice, the expertise of the patent attorney therefore always remains the starting point: we apply a human-driven approach in which AI is used as a supporting tool. No matter how advanced an AI system may be, it functions solely as a fast and intelligent assistant for research and text analysis. AI-generated output is never adopted without scrutiny. Wherever AI is applied, the text is always subjected to human review and editorial control. The substantive choices and the final wording remain entirely the responsibility of the patent attorney.

Moreover, we are fully aware that AI models can never be considered infallible or fully reliable sources of information. Because factual and technical accuracy are crucial in our work, suggestions generated by AI are regarded as valuable input for the development of ideas. However, if an attorney considers adopting such a suggestion, it is always rigorously verified against primary, original sources. The patent attorney provides the indispensable added value: strategic insight, critical judgment, and legal expertise. AI supports the quality of the output by enabling review and serving as a sparring partner for the attorney. The attorney is and remains in control, bearing ultimate human responsibility at all times for both the thought process and the advice delivered.

This cautious and well-considered integration of AI naturally requires robust safeguards, especially in a field where confidentiality is paramount. The instructions and preferences of the client — including with regard to the use of AI in substantive tasks — always remain leading. We are vigilant at all times in protecting sensitive business information and unpublished inventions, and we use only systems that provide appropriate confidentiality safeguards. All our internal guidelines regarding data security and ethical use are aimed at minimizing risks. The EU AI Act (Regulation (EU) 2024/1689) and the AI guidelines of the Institute of Professional Representatives before the European Patent Office (epi) provide valuable professional guidance for the way in which we assess the use of AI.

The key to the future of patent practice lies in finding the right balance: proactively deploying technology as a tool, without detracting from the human thinking, writing, and assessment processes that form the core of our services. In this way, we harness the possibilities of AI where it demonstrably adds value, while confidentiality, client interests, and the professional judgment of the patent attorney always remain paramount.

Want to learn more?

Read the dossier on Patents and AI Inventions here.