Use current LLMs in the notary domain only as supervised support tools in preliminary drafting or educational contexts¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article recommends deploying out-of-the-box LLMs in notarial work only in supporting roles, given their failure to produce valid deeds independently. "These suggests that current LLMs may be useful in the notary domain as support tools in preliminary drafting stages or in educational contexts, provided that their outputs remain subject to strict human supervision." The recommendation rests on the observed residual strengths: most models identified some relevant legal institutes, recognised basic parties' interests, and maintained acceptable linguistic clarity.

Design Implications¶

Context¶

Requirements¶

• Outputs must remain subject to strict human supervision

Constraints¶

• Applies to the notary domain, where all tested out-of-the-box models failed both evaluation steps on complex exam scenarios

Target Learners¶

• notaries and legal professionals
• law students in educational contexts

Target Learning Goals¶

• preliminary drafting of notarial acts
• learning legal institutes and parties' interests

Claims¶

Related Strategies¶

• Use LLM persona simulation only for preliminary experimentation, stress testing, and hypothesis generation, with separate validation for consequential applications
• Use reward-aligned LLM scoring as a teacher-supervised auxiliary tool for preliminary essay scoring at scale

Examples¶

-

Key Sources¶

• Germana Bertoli, Ilaria Amelia Caggiano, Francesca Lagioia, Riccardo Rovatti, Giovanni Sartor, Emiliano Troisi. (2025). What out-of-the-box LLMs can(t) do in law? A Turing test in Italian exams for lawyers, judges and notaries. https://github.com/SSIGPRO/LLMs-in-legal-professions