What should I do first when I find a deepfake?

Document everything before it disappears: screenshot or screen-record the content, the URL, the date and time, and any associated accounts. Don't interact with or report the content yet, because platforms can remove it quickly and you'd lose proof that it existed.

Should I publicly say the deepfake is fake?

Have a brief, clear response statement prepared and endorsed by the right person, but don't scramble to fight the content publicly. Engaging with it can draw attention, increase its reach, and come across as unverified denial rather than a confident rebuttal.

Can I report a deepfake on behalf of someone else?

Yes. If you're an EA, comms director, security officer, or family member, you can engage a removal service directly without waiting for the affected person to act. Getting professional help early matters most, since deepfakes become much harder to contain after the first 48 to 72 hours.

I just discovered a harmful deepfake online, what do I do?

If you’ve just found a harmful deepfake, act fast: document everything before it disappears, don’t engage with or report the content yourself yet, and bring in a professional detection-and-removal service early, because the first 48 hours decide how far it spreads.

Maybe you found it yourself. Maybe a colleague sent you a link with a “did you see this?” message. Maybe your security team flagged it, or a journalist called asking for comment on a video you’ve never seen before.

However you got here, you’re in the right place.

This guide is for anyone who has just discovered that a deepfake is circulating of themselves, or of someone they work for or protect. It’s also for those who haven’t been targeted yet but want to be prepared before it’s too late, because in our experience, the organizations that weather these attacks best are the ones that had a plan and a system in place before it happened. If you didn’t, that’s okay too. You can still stop this. We’ll walk you through what to do, what not to do, and how to move fast, because with deepfakes, speed matters.

First: Don’t Panic, But Don’t Wait

Deepfakes spread quickly. The longer a fabricated video, audio clip, or image stays online, the more it gets shared, downloaded, and re-uploaded to new platforms. Within hours, content that started on one obscure forum can appear on mainstream social media, news aggregators, and messaging apps.

The goal in the first few hours is not to solve everything. It’s to understand what you’re dealing with and get the right people and technology moving.

Step 1: Document Everything Before It Disappears

Before you do anything else, capture evidence.

Screenshot or screen-record the content and the URL where it appears. Note the date, time, and platform. Save any usernames, accounts, or pages associated with it. Do not interact with the content yet, no comments, no reports. We’ll get to that.

Platforms sometimes remove content quickly, which sounds good, but if you haven’t documented it first, you lose proof that it existed. That proof matters for platform appeals and law enforcement if it comes to that.

Step 2: Assess the Threat

Not all deepfakes are the same. Ask yourself, or the person you’re supporting, a few quick questions:

What type of content is it? A fabricated video of someone saying something false? A voice clone used in a fraud call? A manipulated image? Each requires a different response.

Where is it circulating? A closed Telegram group is different from an open Instagram account.

What is the intent? Reputational damage? Financial fraud? Political manipulation? Harassment?

How widely has it spread? Is this one post, or is it already being reshared?

The answers shape how urgently you need to act and what channels to prioritize.

Step 3: Prepare a Response Statement

Before anything goes public, prepare a brief, clear statement. This doesn’t need to be released immediately, but having it ready matters. A short message confirming the content is fabricated, endorsed by whoever is appropriate in your organization, gives you something to act on the moment a journalist, partner, or stakeholder reaches out. Scrambling to write a response under pressure is avoidable, and it shows.

Step 4: Do Not Try to Handle This Alone

This is the mistake most people make. They report the content to the platform, get an automated rejection, and lose days waiting for a second review. Meanwhile the content spreads.

Platform content moderation systems are not built for deepfake cases. Standard reporting queues are slow, inconsistently enforced, and often staffed by reviewers who don’t have the tools to verify that content is AI-generated. A general “report” button is not a takedown strategy.

What actually works is a combination of verified detection, confirming and documenting that the content is artificially generated with technical evidence, targeted removal using the right legal and technical levers on each specific platform, and continuous monitoring to track whether the content re-emerges after removal, which it often does. This is why detection and removal have to work together rather than as separate steps.

Step 5: Get Professional Help Early

The earlier a professional service is involved, the better the outcome. Deepfakes that spread for 48 to 72 hours become significantly harder to contain as copies proliferate across platforms, caches, and private channels. Some platforms have escalated response channels that are only accessible to verified removal partners, not to individuals filing standard reports.

If you’re an EA, comms director, security officer, or family member acting on someone else’s behalf: you can engage a removal service directly. (If the target is a leader at your company, this guide on how to report and remove a deepfake of your executive walks through the specifics.) You don’t need to wait for the affected person to take action themselves, especially if they’re in the public eye and the situation is sensitive.

What Revelum Does

Revelum is a deepfake detection and removal service. When you engage us, we run a full detection scan, including biometric analysis and platform-wide monitoring, to understand the full scope of what’s out there. We initiate targeted removal across the platforms where content is identified, monitor for re-emergence after takedown, and apply proprietary measures designed to dismantle the infrastructure behind the attacks. (Here’s what happens after a deepfake is detected, step by step.) We also provide documentation of findings and actions taken, which can support any formal proceedings if needed.

This is what we do, and we’ve been doing it long enough to have built the data and operational experience that makes a real difference when it counts.

We work with individuals, executives, public figures, and organizations across the Americas and Europe.

The Bottom Line

The best time to prepare for a deepfake attack is before it happens. The second best time is right now.

If you’ve already found something, document it, don’t engage with it, and contact a professional service that can move fast and knows how these systems work. If you haven’t been targeted yet, consider what a plan would look like before you need one.

Contact Revelum

We’ll assess your situation and tell you exactly what we’re seeing and what can be done, typically within 24 hours.

Revelum is a deepfake detection and removal service operating globally, with a focus on the Americas and Europe. We protect executives, public figures, political leaders, and organizations from AI-generated disinformation and fraud.

Disclaimer: This article is intended for informational purposes only and does not constitute legal advice. Every situation is different, and we strongly recommend consulting a qualified legal professional for guidance specific to your circumstances. Revelum’s services are operational in nature and do not replace legal counsel.