Marco Giacomello.Mens et opusITWork with me

Deepfakes become a crime

25 September 2026 · 1 min

Italian Law no. 132 of 23 September 2025 added Article 612-quater to the Criminal Code: "Unlawful dissemination of content generated or altered with artificial intelligence systems".

It punishes anyone who causes unjust harm to a person by transferring, publishing or otherwise disseminating, without their consent, images, videos or voices falsified or altered with these systems and capable of misleading as to their authenticity. The penalty is imprisonment from one to five years.

Three points to note

First: there must be unjust harm. Creating the content is not enough; what matters is the dissemination that causes harm.

Second: the offence is prosecuted on complaint by the injured party. It is prosecuted ex officio if connected with another offence prosecuted ex officio, or if committed against a person who is incapacitated by age or illness, or against a public authority because of its functions.

Third: the same law introduces a new general aggravating circumstance, in Article 61 no. 11-decies. It covers those who commit an offence using these systems when, by their nature or manner of use, they constitute an insidious means, hinder defence or aggravate the consequences.

A criminal provision alone is not enough. We need tools to recognise fake content and fast procedures to take it down. But the message is clear: a person's voice and face are not free raw material.

Let’s discuss

Is a criminal provision enough to protect people's voices and faces, or do we mainly need tools to remove fake content quickly?

Reply by email

Personal reflection, not legal advice.

Newsletter

Two ideas a week, straight to your inbox

  • Notes on law, mind and technology
  • Open challenges, before they hit the site
  • Early access to new products

← All writing