On May 5, 2026, Italy’s prime minister, Giorgia Meloni, publicly addressed the spread of manipulated photographs created with artificial intelligence. She shared a sample image that had been widely distributed online and explained that the picture was a fabrication. The episode highlights how easily a single deepfake can gain traction and why public figures — and ordinary people — must be alert to deceptive content. Meloni framed the incident as more than a personal affront: she warned that such technologies can be weaponized to mislead and to damage reputations.
In a brief public message, Meloni noted that several images of her had been generated using AI and circulated as if authentic by political adversaries. She even made light of the alteration, saying the creator had “improved” her appearance, but she immediately turned the tone serious, calling the tactic “cyberbullying” and urging users to exercise caution. Her post included a fabricated picture that appeared to show her seated on a bed in lingerie, an image that spread rapidly and provoked angry reactions from people who believed it was genuine.
How the incident unfolded
The manipulated image moved quickly across social networks, shared and reshared without verification. Observers criticized those who treated the photo as factual; one comment labeled such images as “shameful” and unworthy of the office of the prime minister. The rapid circulation demonstrates the potency of deepfakes as a misinformation vector: a single synthetic file can create an illusion of truth that is difficult to undo. Experts describe a deepfake as an AI-generated alteration that replaces or modifies a person’s likeness in still images or video, often with convincing results.
Why this matters beyond a single image
Meloni warned that the risks extend far beyond political theatre. The use of artificial intelligence to produce manipulated content can mislead the public, stoke harassment, and target individuals who lack resources to defend themselves. She emphasized that while she can respond publicly, many victims of non-consensual manipulations cannot. Her message boiled down to two practical injunctions she urged the public to adopt: verify before believing and think before sharing, a reminder that digital literacy is a frontline defense against manufactured falsehoods.
Legal and institutional responses
Legislation and enforcement
The episode comes against a backdrop of intensified policy moves in Italy. The government has made tackling the harms of AI and deepfakes a priority, advancing laws that impose penalties for malicious manipulations and setting limits on certain uses of the technology. Officials say these measures are intended to align with broader European rules while giving national authorities tools to prosecute offenders who distribute harmful, manipulated imagery. Enforcement has included investigations and takedowns in high-profile cases where images were used to humiliate or extort.
Past incidents and investigations
Italian authorities previously confronted a scandal when a pornographic platform published doctored images of prominent women, lifted from social media and altered with explicit captions. That case prompted police orders to shut the site and prosecutors to open inquiries into suspected offenses such as unlawful dissemination of explicit images, defamation, and extortion. Meloni herself launched a libel suit two years ago against an individual accused of creating and sharing fake pornographic images using her likeness, underscoring that legal recourse is already part of the response mix.
What individuals and platforms can do
Combating the spread of deceptive material requires both personal caution and structural fixes. On a personal level, Meloni’s advice to the public—verify before believing and think before sharing—is practical: check original sources, reverse-image search suspicious photos, and consider whether a piece of content seems engineered to provoke. Platforms and policymakers must also improve detection, enforce transparency rules, and create clear reporting channels for victims. Together, these steps can reduce the efficacy of deepfakes and protect those who are most vulnerable to digital exploitation.
