Insights

How Many Lawyers Use AI in Italy: CENSIS Data

3 min read · Editorial oversight: Avv. Federico Papa

The adoption of artificial intelligence (AI) in the Italian legal profession has seen unprecedented growth in recent years, radically transforming the daily practice of law firms. According to official data, the profession is divided between those embracing technological innovation and those fearing risks related to confidentiality and data accuracy. Choosing the right tool has become the dividing line between professional efficiency and the risk of disciplinary and procedural sanctions.

In brief

The CENSIS and Cassa Forense report indicates that 27.5% of Italian lawyers utilize AI, with high adoption rates among professionals under 40. Pursuant to the Italian Code of Forensic Deontology (Articles 12 and 27) and CNF guidelines, lawyers must provide prior disclosure to clients regarding AI usage. Improper use of generalist models leads to aggravated procedural liability under Article 96 of the Code of Civil Procedure and sanctions from the Data Protection Authority for lacking a DPIA. Human intellectual input remains irreplaceable, prohibiting total delegation to algorithms.

1.
Evolution of CENSIS data on AI adoption
According to the recent Report on the Legal Profession prepared by CENSIS and Cassa Forense, initially 27.5% of Italian lawyers used AI, with an additional 31.7% intending to do so. However, the most recent survey data show a historic turning point: an increasing number of lawyers now rely on AI, reaching significant peaks among professionals under 40.
2.
Duty of disclosure and transparency toward the client
In accordance with the duties of information and competence (Articles 12 and 27) and CNF guidelines, the lawyer is required to inform the client in advance about the use of AI systems. This rule aims to protect the fiduciary relationship, ensuring that the client is fully aware of the deployment of technological tools in the performance of the legal mandate.
3.
Risk of sanctions for using generalist AI
The use of generalist AI models has led to severe procedural consequences, including sanctions imposed pursuant to Article 96 of the Italian Code of Civil Procedure. An emblematic scenario involves aggravated procedural liability resulting from the improper use of unverified tools that may generate inaccurate content or non-existent legal citations.
Apply this research directly with edit.legal
Legal research and drafting with citations checked against official databases. Try edit.legal for free, no credit card.
Try edit.legal
4.
CNF guidelines and professional liability
The Italian National Bar Council (CNF), adopting the CCBE guidelines, reaffirmed that human intellectual contribution must remain predominant. The lawyer retains full responsibility for every legal document produced, with an absolute prohibition on total delegation to algorithms, particularly to prevent the so-called hallucinations typical of generalist models.
5.
Privacy and compliance with the Data Protection Authority
The Italian Data Protection Authority has intensified checks on AI tools that lack an adequate Data Protection Impact Assessment (DPIA). Using systems that fail to guarantee data confidentiality and professional secrecy exposes lawyers to sanctions for breaching privacy regulations.
6.
The edit.legal solution: verified sources and control
To meet these challenges, edit.legal provides a platform grounded in over one million verified documents, including legislation, Supreme Court case law, and the Official Gazette. Based on the analysis of thousands of real legal queries, we developed 21 practice areas in which multiple AI agents collaborate on each case to ensure verified citations and absolute precision.
7.
Technical integration and European infrastructure
The edit.legal ecosystem includes a native legal editor with variable fields, an add-in for Microsoft Word, and MCP servers for integration with external tools. The entire infrastructure resides on GDPR-compliant European servers, offering transparent pricing and a free trial with no credit card required or mandatory demo.

Editorial oversight: Avv. Federico Papa·ICAM

Frequently asked questions

What are the risks for a lawyer using ChatGPT or other generalist models?

The main risks concern the generation of hallucinations and non-existent legal citations, which can lead to professional and procedural liability, and privacy breaches. Generalist tools often lack an adequate DPIA, exposing lawyers to enforcement actions by the Data Protection Authority.

What do disclosure duties require regarding the use of AI?

Duties of diligence and ethical rules require the lawyer to inform the client of the use of AI systems in performing the legal mandate. This obligation serves to ensure transparency and safeguard the fiduciary relationship between lawyer and client.

Is it possible to entirely delegate the drafting of legal documents to AI?

No. Pursuant to CNF guidelines and legal profession regulations, human intellectual input must remain irreplaceable. The lawyer retains sole responsibility for the contents produced and cannot abdicate their analytical role to an algorithm.