Here is the rewritten text:

Scammers are increasingly using synthetic voices and deepfakes to deceive victims in Brazil, making it harder to identify the fraud. According to Ubiratan Cascales, a threat intelligence expert, these criminals are exploiting Generative AI technologies to clone faces and voices, even fooling close relatives and large companies.

“The Brazilian scammer is one of the most creative in the world. We’ve already seen cases where people lost money after seeing or hearing a family member in a video call, but it was all fake, created with AI,” explains the expert.

What are synthetic voices and deepfakes?

Synthetic voices are audio files created by AI from real speech fragments of someone. With just a few seconds of audio, it’s possible to generate complete speeches with similar tone and emotion to the original person.

Deepfakes are manipulated videos that use real images to create extremely realistic faces and expressions. This technique, once used in cinema, is now accessible to anyone on the internet for free.

“With these tools, anyone can recreate someone else’s voice and face and impersonate them. It’s enough to type a text, and the AI transforms it into a convincing video,” warns Ubiratan.

Most common scams in Brazil

According to Ubiratan, the most common scams in Brazil involve:

• Phone calls or messages with cloned voices of sons, parents, or spouses;
• Fake videos with executives’ faces and voices authorizing bank transfers;
• Emotional manipulation scams, where deepfakes are used to deceive victims;
• Fake gift or prize scams, where criminals collect images or audio from victims to feed AI systems.

How to protect yourself from scams?

Ubiratan shared some practical tips to protect yourself from these technological frauds:

• Distrust urgent money requests, even if they seem to come from known people.
• Hang up and try to confirm the contact through a saved number.
• Keep your social media profiles private, if possible. The less public information, the harder it is for scammers to set up a scam.
• Limit who can see your WhatsApp profile picture. Configure it to only allow contacts to access.
• Be cautious with online videos and audio files, especially those with your voice. They can be used to train AI systems.

“If someone asks you for money urgently, there’s a 99% chance it’s a scam. Take a deep breath, hang up, and confirm with calm. Emotion is the scammer’s weapon,” he emphasizes.

Most affected sectors:

The financial sector is the most impacted by this type of fraud, according to the expert. However, no one is immune: companies, governments, and even small businesses are targets for criminals with access to AI tools.

“Even a restaurant can be a victim. The scammer doesn’t need hacking, they use social engineering and emotional manipulation to get what they want,” he affirms.

While banks and large companies are investing in security, the public sector is still lagging behind in combating these crimes. According to Ubiratan, it’s essential for companies to adopt a preventive approach, focusing on predictive intelligence: studying threats before they become attacks.

“The use of AI by cybercriminals is already a reality in Brazil. And the best way to protect yourself is with knowledge, attention, and validation of information. At the slightest sign of doubt, don’t act impulsively. Confirm with calm and seek help,” concludes the expert.

Want to hear the full interview?

The episode with Ubiratan Cascales is available on the podcast, on major audio platforms. Click here and check it out!