Uncovai

Quick tip: you don't have to become a detective to stay safe online🕵🏻‍♂️ Yes, you can look for odd fingers, strange blinking or skin that looks too smooth. But AI keeps getting better at hiding those signs, and nobody has time to inspect every photo they receive.

An easier habit: before you share or believe something, run it through https://uncovai.com
➡️ Drop the photo, video or audio in, and you'll see in seconds whether AI is likely behind it. Just a quick, simple signal that helps you decide. Try your first month free!

🔗Android App: https://play.google.com/store/apps/details?id=com.uncovai.app
🔗iOS App: https://apps.apple.com/fr/app/uncovai-ai-detector/id6809812427?l=en-GB

How can we detect AI-generated and AI-modified images, and where does detection fall short?

As AI-generated and AI-modified imagery becomes increasingly sophisticated, determining what is real, what has been altered and how it has been altered is a growing challenge for the visual media and news industries.

For our next CEPIC webinar, we’ll be joined by Florian Barbaro, CEO of Uncovai, who will share his expertise in AI detection, alongside Mete Zihnioglu, Directeur Général of Sipa Press, who will offer the perspective and experience of an independent press agency working with press photography.

We’ll explore what today’s AI image detection tools can identify, where their limitations lie, and what happens when images are generated, retouched, cropped, edited or modified using AI.

CEPIC members - we need your help!

We’re looking for real-world images to use as practical examples during the webinar. These could include:

• Original, unmodified images
• Photoshop or otherwise retouched images
• Fully or partially AI-generated or AI-modified images
• Cropped or resized images
• Images that have undergone several different modifications

If you would like to contribute, please get in touch. Your examples will help us put AI detection to the test and demonstrate both its possibilities and its limitations.

Save the Date:
📅 Tuesday 20 October
🕔 17:00 CET

Registration details will follow soon.

‼️More than half of the political images on X are now AI-generated‼️

We analyzed thousands of posts ahead of the 2026 US midterms🇺🇸

In our sample:
🔹 55% of images were AI-generated, 45% were real
🔹 Pro-Trump voices led in both volume and engagement
Synthetic media is no longer a fringe in political feeds. It's the majority of what people scroll past.

Together with Laboratoire SAMM (University of Paris I: Panthéon-Sorbonne), we put the data, the methodology and what to watch before November into two free reports.

Link in the first comment 👇

An Italian bank just lost €95 million, and it took one phone call 📞
The voice on the line sounded exactly like a lawyer the chairman knew. It was AI-cloned. A fake WhatsApp from “the CEO” set it up, and a spoofed email supplied the account details, nobody hacked a single system.
⠀
Voice cloning needs only a short audio sample, and it can happen to anyone, not only banks.
⠀
Here’s how UncovAI secures your communications:
⠀
📞 Zoom and Teams: real-time detection of cloned voices and deepfake video
💬 WhatsApp and Telegram: bots that verify suspicious messages, photos, videos or voice notes
✉️ Outlook: phishing link detection before anyone clicks
⠀
Verify first by one click!
🔗 Free for one month! https://uncovai.com
Android App - https://play.google.com/store/apps/details?id=com.uncovai.app

Which side is your idol on?🔎

UncovAI — spot fakes in just one click. Free for one month!
https://uncovai.com
Android App - https://play.google.com/store/apps/details?id=com.uncovai.app