Media headlines often focus on the incredibly realistic results produced by AI image generation tools, but they rarely highlight how those deepfakes are actually created and at what cost.

Many deepfakes are built from ordinary photos, videos, and voice recordings — the everyday content that we willingly upload and make publicly available on social media like Instagram or Facebook, and cloud storage like Google Drive or OneDrive.

What we rarely consider is that those digital traces can be reused in ways we never intended, including for creating deepfakes without our knowledge or consent. This usually means losing privacy and control over how we appear online, which can lead to scams, identity theft, harassment, cyberbullying, reputational damage, and emotional distress — often irreversible(nieuw venster).

Non-consensual deepfake imagery like the Sweet Anita case has become a serious form of abuse. In the US, the DEFIANCE Act(nieuw venster) aims to strengthen legal recourse for victims, and the Take It Down Act(nieuw venster) focuses on the faster removal of harmful content online.

In the EU, reports of Grok being used on X to generate sexualized deepfake images (including of women and minors) have prompted privacy regulators to investigate(nieuw venster) whether this could violate GDPR data protection laws. At the same time, countries like Spain(nieuw venster) are moving ahead with national legislation to criminalize non-consensual deepfakes.

Until laws catch up with how fast deepfake technology is evolving (often to the point where it’s hard to tell what’s real), it’s important to be proactive about protecting your photos, videos, and voice recordings. Preventing misuse is much easier than dealing with the fallout later.

How to prevent deepfakes

Deepfakes depend on three things: the quality of the data, how much data is available, and how easy it is to access. Here’s how to protect yourself against deepfakes:

how to prevent deepfakes tips

Be selective about what you share publicly

The more high-resolution content is available, the easier realistic cloning becomes. Here’s how to reduce how much high-quality media of you is publicly available:

  • Avoid posting HD close-ups of your face or long clear videos, especially those with consistent angles and lighting. 
  • Remove old public media you don’t need online anymore.
  • Separate public and private accounts as much as possible.
  • Ask others not to tag or upload media of you without consent.
  • Review app permissions and remove unused third-party apps that may have access to your camera roll.

Make your public photos and videos harder to reuse

When sharing photos or videos publicly, there are ways to make them harder for AI systems to copy, analyze, and reuse them. These methods don’t completely prevent deepfakes or AI training, but they can discourage misuse:

Adjust your privacy controls on social media

Many of these controls are easy to overlook but can make a real difference in limiting how much of your content is visible to strangers and data scrapers:

  • Set your accounts to private where possible.
  • Limit who can see past and future posts (friends-only instead of public).
  • Restrict who can download, share, or repost your content.
  • Turn off facial recognition features if the platform offers them.
  • Control tagging settings so you must approve posts you’re tagged in.
  • Limit who can find you using your email or phone number.

Remember that the platform itself may still have access to your data and process it under its own terms. That’s why it’s worth reviewing privacy and AI usage policies from time to time. If you’re uncomfortable with how much of your data may be used, consider limiting what you upload, or whether you use that platform at all.

For example, Meta may use all Meta AI interactions to create targeted ads and improve its AI systems. So content you share in what feels like a private conversation with a friend — like attaching an image of yourself and asking Meta AI to generate a funny version of it — may end up in an AI dataset where you have no control over what happens to your image.

Choose tools that respect your privacy

Some apps are designed to collect as little personal data as possible and give you control over how your content is stored and shared — an important part of digital sovereignty. No single tool can completely stop deepfakes, as it largely depends on what you choose to share about yourself. But privacy-focused services like those in the Proton ecosystem can help limit how much of your photos, videos, and voice recordings end up publicly accessible or available for AI training.

Proton Drive is a free, secure cloud storage and file sharing service that uses end-to-end encryption. This means only you and the people you explicitly share files with can access your data, not even Proton. You can share files, photos, or albums through restricted email invites for smaller groups, or through secure public links that can be password-protected and set to expire automatically. Proton Drive lets you easily track downloads and revoke access anytime.

Start with 5 GB free

Check your digital footprint from time to time

Your digital footprint grows passively over time. Even if you’re careful now, older posts, recordings, tagged photos, or archived media can resurface and remain publicly accessible without you noticing. Plus, platforms may change default privacy settings.

Every few months, you should check what’s available of yourself online that could be used to create deepfakes. Here’s how:

  • Search your name and images on search engines periodically.
  • Review the visibility of your posts.
  • Review follower or contact lists for unfamiliar or suspicious accounts.
  • Remove outdated content or restrict it to smaller audiences.
  • Close unused accounts.
  • Recheck privacy defaults and AI training settings after platform updates.

Deepfake prevention tips for families

It takes only 20 photos or a 30-second video to create a realistic digital profile of someone. That means everyday family photos, school videos, voice notes, or social media posts can unintentionally become training material for deepfake tools.

For families, deepfake prevention is less about strict control or fear, and more about awareness, communication, and thoughtful sharing habits — especially for children who can’t fully understand long-term online risks yet.

Helping kids develop good digital instincts early, like asking before posting, thinking about audience, and understanding how biometric data can be reused, can make a lasting difference. And for adults, being intentional about what gets shared publicly versus privately helps reduce unnecessary exposure while still preserving memories.

Here are some deepfake prevention techniques that families should keep in mind:

For parents

  • Avoid sharenting, especially public albums, milestone posts, or identifiable school and location details.
  • Children’s faces change as they grow, but old images can still be used to train AI models, so it helps to reduce public photo archives.
  • Use private sharing channels, such as family chats on Signal or private albums on Proton Drive, instead of public social media.
  • Regularly review privacy settings and tagging permissions on your own accounts.
  • Talk openly with relatives, friends, and educators about not posting or reposting your child publicly without permission.
  • Keep an eye on where others might be sharing photos — like schools, clubs, sports teams, or photographers — and opt out when you can.

For kids and teens

  • Think before posting: would you be okay if this photo or video spread beyond your friends?
  • Be cautious with viral challenges, face filter apps, or AI avatar apps.
  • Don’t share personal details (including school, address, routines) alongside photos and videos.
  • Ask friends before posting them, and expect the same respect.
  • Understand that deleting something doesn’t always remove copies. It’s safer to avoid posting it in the first place.
  • If something feels uncomfortable online, trust that instinct. Tell a parent or trusted adult if someone asks for photos, voice clips, or unusual content.

Stay in control of your digital identity

Deepfakes are reshaping how identity, privacy, and trust in people and companies work online. While laws and platform policies are evolving, they often move slower than the technology itself, which makes awareness and proactive digital habits more important than ever, both for you and for the people around you.

Being thoughtful about what you share, where you store it, and who can access it won’t eliminate risk entirely, but it does make a difference. That focus on control and privacy is at the core of Proton’s mission: building tools that help people stay in charge of their data online.