NLO's interpretation of the Future of Patent Law Integrating and Generative AI

Highlights

· The epi Guidelines provide guidance for patent attorneys on effectively and ethically integrating generative AI into their daily practice while maintaining the highest standards of professional conduct and client service.

· Patent attorneys must actively seek to understand the capabilities, limitations, and confidentiality features of AI models, ensure transparent communication with clients, and maintain accountability for the quality and accuracy of AI-generated work.

· As the legal landscape evolves, patent attorneys must stay informed about new legislation, such as the European Artificial Intelligence Act 2024, and adapt their practices to ensure compliance with professional obligations and client expectations.

The European Patent Institute (epi) has recently published Guidelines for the use of Generative AI in the Work of Patent Attorneys. These guidelines come at a key time, as the rapid progress of AI technology is changing the way patent professionals operate. By following these guidelines, patent attorneys can use AI while maintaining high standards of professional conduct and client service.

Understanding AI Models

One of the main ideas in the epi Guidelines is the importance of understanding the abilities and limits of generative AI models. As the guidelines point out, "It is however an essential characteristic of professionalism that Members do their utmost to inform themselves about the weaknesses of any generative AI model used in their professional work." This active approach to learning about AI technology is to make sure that attorneys will know how to use AI to assist in tasks where it will actually make a useful contribution, and not in tasks for which the AI is poorly prepared.

In particular, the epi Guidelines warn against the risk of "hallucinations" – wrong or misleading information presented as fact by AI models. As Guideline 1 cautions; "A common weakness is a so-called hallucination, i.e., an AI generated response which contains false or misleading information presented as fact." By staying informed about the specific attributes of the AI models they use, patent attorneys can lower these risks and make good choices about how they use AI.

Also, the guidelines talk about how patent attorneys need to look into the confidentiality features of AI models. As stated in Guideline 2a; "Members should take active steps to establish whether a chosen AI model assures the confidentiality of material fed to it." This also ties in with understanding how the AI works – the attorney must be aware of the terms and conditions of the AI product, in particular what will be done with text input and output, and the jurisdictions that apply. Only when armed with this knowledge can the attorney make an assessment whether or not the requirements of client confidentiality, for which the attorney carries a legal responsibility, are adequately met.

Harm van der Heijden from the NLO AI Focus Group states: “The EPI guidelines in particular note a risk that AI models with a memory of past interactions can potentially spread confidential information. If an attorney uses a model with such a memory feature to draft a patent application for client A, and later uses that same model to draft an application in a related field for client B, some of the model's memories that are confidential to client A might be injected into the material for client B. You might say that the AI model has a conflict of interest in this case! Joking aside, it is absolutely essential that the attorney is aware of such possible "contagion". In Guideline 6, the EPI recommends to use separate user accounts for such AI models. While this may be somewhat of an excessive measure (there are generally easier ways to ensure that an AI model starts from a clean slate), it shows that the EPI has thought deeply about possible problems.”

Keeping Professional Responsibilities

Using AI in patent work does not in any way reduce the professional duties of patent attorneys. As the epi Guideline 3a clearly state; "Members remain at all times responsible for their professional work, and cannot cite the use of generative AI as any excuse for errors or omissions." This is an important (and hopefully unnecessary) reminder that the final responsibility for the quality and accuracy of patent work belongs to the attorney, not the AI tool.

To make sure AI-generated content meets the same high standards as work done by people, patent attorneys must develop strong processes for reviewing and checking this content. The guidelines stress this, stating, "Members must check any work product produced using generative AI for errors and omissions. The checking process must ensure that the work product is at least of the same standard as if it had been produced by a competent human practitioner." [see Guideline 3b]

Van der Heijden: “This part of the guideline is of the utmost importance. As any person experienced with AI can tell you, it can be quite difficult to recognize poor work from an AI because superficially, all text generated by AI will typically look authoritative and convincing, even though in reality it may be complete nonsense or, more commonly and perhaps more dangerously, form a suboptimal solution to the client’s requirements. Especially clients who are not attorneys will have difficulty assessing the quality of AI-generated work. That is why it is of the utmost importance that the attorney remains in control of the process, only uses AI tools to assist in the implementation of the solution that the attorney has personally designed, and rigorously checks all parts of the result involving AI contributions.”

Also, the epi Guidelines emphasize putting clients' interests first and maintaining the highest levels of professional conduct. As the guidelines state, "When using AI of any kind in professional work, a Member must adopt the highest possible standards of probity; must take all reasonable steps to maintain confidentiality when this is required; and at all times must put the interests of clients first as required by Article 1 of the epi Code of Conduct."

Communicating AI Use

Being transparent is a key idea in the epi Guidelines, especially when it comes to telling clients about AI use. As the guidelines overarching principles note, "Members cannot assume knowledge of the wishes of clients with regard to generative AI. Some clients may encourage the use of generative AI, whereas some others may oppose its use in their cases." This requires attorneys to actively communicate and get informed consent from clients about possibly using AI in their matters.

To help with this communication, Guideline 4 suggests that "Members keep accurate records of enquiries sent to clients in order to establish their preferences with regard to generative AI, and the responses to those enquiries." By keeping detailed records of client preferences and communications, patent attorneys can make sure their use of AI matches what their clients expect and want.

Van der Heijden: “At the moment, this is mostly relevant for translation services, because there the use of AI-based machine translations is rapidly increasing. NLO always informs their clients if translations are based on AI generated translations, even when followed by a human check.”

Staying Up to Date on Laws

As AI technology keeps advancing, so will the laws around its use. The epi Guidelines highlight the importance of staying informed about relevant laws, like the upcoming European Artificial Intelligence Act 2024, which will take effect in January 2025. Guideline 7a states; "At the very least, EU-based Members must have regard to, and apply the provisions of, the European Artificial Intelligence Act 2024 (Regulation (EU) 2024/1689) insofar as these impact practices." Patent attorneys are already used to staying up to date with changing patent laws, case law concerning patents, laws concerning other intellectual properties, privacy laws and general civil law. Now they can add AI laws to this list as well.

In addition to legal obligations, patent attorneys must also know about any extra restrictions, duties, or reporting requirements from external groups, like national patent attorney associations or professional insurance providers. And Guideline 7b warns; "Members should in addition have regard to any restrictions, obligations or reporting requirements, imposed by external organisations, that may impact the extent to which or the ways in which generative AI models may be used."

Setting Fees in the Age of AI

Using AI tools also raises important questions about how fees are set for patent attorney services. The epi Guideline 8 gives advice on this issue, stressing the need for fairness and transparency. As stated, "When determining fees to be charged for work products generated using AI tools, Members must at most charge fees that fairly reflect the amount of time and effort and/or the degree of difficulty and/or the degree of risk involved." In other words, when the use of AI results in an increase in efficiency, the clients should also benefit. Clients are not to be overcharged for AI-generated work and that fees should accurately show the value provided by the attorney.

Also, this Guideline notes that "Members may charge, at levels fairly reflecting the difficulty or extent of the task, for setting up or training of AI tools, AI tool subscription fees and for checking AI-generated work." This recognizes the costs and work that go into bringing AI into practice and lets patent attorneys include these expenses in their fee calculations. However, the guidelines stress being transparent, suggesting that "Members keep accurate records of all aspects of setting up, training and checking work, including the levels of experience or professional education of those responsible for achieving an accurate output." By keeping detailed records and openly communicating with clients about how AI impacts fees, patent attorneys can build trust and make sure using AI in practice is sustainable over the long term.

In summary: the epi Guidelines on using Generative AI in

The epi Guidelines on using Generative AI in patent attorney work provide a thorough framework for using this transformative technology in the context of a patent attorney’s practice. By understanding AI models, keeping professional responsibilities, being transparent about AI use, staying informed about laws, and setting fees fairly, patent attorneys can use AI to improve their services and better serve their clients.