Responsible Usage of AI by Legal Professionals

According to a 2023 study by Goldman Sachs, almost half of all legal tasks could be automated. This poses major challenges for responsible managers of law firms and legal departments: How can and must they change and transform to remain well positioned as knowledge workers in the future?

Strategy more important than tools
It is tempting to get bogged down in details too early on: What are the best use cases, which tools should be procured, how much will it all cost, and how do you prompt correctly? These questions are certainly important, but they are operational in nature and are addressed more downstream. Managing trends and challenges primarily means clarifying the fundamental questions: What risks can be expected and how do we want to position ourselves in the new world (see Asterix effect)? What strategy are we pursuing and which business model will be suitable tomorrow (see hourly fee, leverage, and one-stop shop)? How is leadership changing as a key factor? In any case, it is foreseeable that an intensified customer focus and the human element will become more important.

Don’t forget the human element
Experts such as lawyers, doctors, engineers, and architects tend to focus on factual knowledge because it provides them security. And managers love modeling Excel sheets because, ultimately, the numbers matter. But where does that leave the human element? Today, lawyers like to claim that, despite overwhelming technology, human lawyers will still be needed tomorrow. But do they act accordingly?

Because humans will become more important, there are a few points to consider, because humans are considered the weakest factor in the equation.

• Convenience: We like to take the path of least resistance, and inertia protects us from unnecessary waste of resources. However, AI is like sugar: it offers a huge temptation and is available anytime, anywhere, in vast quantities, and practically free of charge. Applied to AI, this means that people like to quickly search for an answer using AI instead of first dealing with the problem intensively.

• Plausibility check: It’s too good to be true. In the future, AI will be able to do our work, and humans will only need to do a sanity check. But how exactly should that work? We will only see what we know and understand (Goethe). Consequently, we must continue to painstakingly train and feed our brains with knowledge and experience in order to be able to correctly assess the output produced by AI.

• Rocket science: Similarly, the desire to leave the tedious groundwork to AI and only engage in rocket science remains an illusion. Without many years of laborious and painful practice, we will never reach world-class status.

• Critical thinking: Humans claim to be capable of critical thinking. But isn’t this too often checked at the wardrobe? Today everyone knows that AI – at least: still – hallucinates. Nevertheless, the results produced by AI are accepted without contradiction and verification. This is demonstrated by the many fake quotes from lawyers and, more recently, from courts, even three years after ChatGPT was introduced.

• Working methods: Working methods and the value chain will change. Today, the following steps are still applied: searching (e.g., on Google), reading the source, writing, and finally active thinking. With AI, this could change to the following: prompting, reading the summary, editing the result if necessary, and passive plausibility checking.

“Just because we have artificial intelligence doesn’t mean we have increased natural intelligence.” (Eric Guyer, NZZ, April 12, 2025). Ensuring digital competence is an ongoing task, even for lawyers. Technology is constantly changing, seemingly at the speed of light. The number, frequency, and magnitude of the changes overwhelm us, are cause for concern. We can no longer absorb everything properly, let alone understand it. While we are trying to understand and digest AI, the next challenge is already waiting around the corner (e.g., quantum technology). Instead of getting lost in details too early, technological changes must be approached strategically and with foresight.

Responsible management requires entrepreneurial thinking. This also means being aware of new challenges, developing appropriate control measures, continuously reviewing them along the way, and ultimately adjusting them where necessary. Is this easy and convenient? Not at all, but it is exciting! In all the enthusiasm for the new, we should not forget the humans who cannot adapt at the same pace.

Got curious? – Curious to learn more? Take a look at the presentation on this topic, which goes into more detail on many other points. It is available in both German and English.