Bart Custers writes blog post on judge who used ChatGPT
This month, it came out that a subdistrict court judge used ChatGPT to formulate a ruling. The judge included some text generated by the popular AI chatbot in his decision. This resulted in a lot of media commotion, raising doubts as to whether using ChatGPT in court is permissible and responsible.
However, this isn’t cause for panic, argued Bart Custers, Professor of Law & Data Science at eLaw, Leiden University’s Center for Law and Digital Technology in a blog for Nederland Rechtstaat (in Dutch), on 2 September 2024. Judges using ChatGPT in their work is not necessarily problematic, provided there are clear ground rules. In fact, as in many other sectors, judges’ work can be supported, facilitated and even improved by AI.
Judges who do everything they can to reach a correct, fair decision demonstrate their position at the centre of society and that they are well informed and carefully justify their decisions. The use of AI in court is predicted to increase over the coming years, so we’d better get used to it, argues Professor Custers.
See the following websites for more information (in Dutch) on related decisions and rulings:
- Rechter gebruikt chatbot ChatGPT in een uitspraak
- Uitspraak rechtbank Gelderland
- Gebruik van ChatGPT door rechter lokt grote discussie uit onder juristen
- Ongeloof om Nederlandse rechter die ChatGPT gebruikt in vonnis
Read Professor Custers' full blog post (in Dutch) here.