safetyAIcrisis

Checklist: What to Do When an AI Tool Generates Harmful Content About You

UUnknown

2026-02-13

11 min read

A concise, prioritized checklist for creators facing non-consensual or sexualised AI content—evidence preservation, takedowns, legal steps, and comms templates.

Hook: When an AI tool makes sexualised or non-consensual content of you — act fast

If an AI tool (for example, recent misuse of Grok) or a social platform has generated sexualised, explicit, or non-consensual images or video of you, this is a crisis that requires a clear, prioritized response. Creators and publishers lose revenue, followers, and trust when harmful AI content spreads — but the right steps taken in the first 24–72 hours greatly increase the chance of removal, legal remedies, and reputation recovery.

Executive summary — most important actions first

Preserve evidence (screenshots, downloads, metadata, hashes, logs).
Report and request takedowns with platforms and host providers using clear templates.
Get legal help and send preservation/cease-and-desist letters when appropriate.
Communicate deliberately with followers, press, and your team — use provided crisis comms templates.
Lock down your accounts and start monitoring to prevent repeat abuse.

Why this matters in 2026

Late 2025 and early 2026 saw attention on AI harassment escalate — regulators and attorneys general opened probes into tools like Grok after reporters were able to generate sexualised deepfakes of real people and post them publicly. Platforms are scrambling to update policies, but enforcement is inconsistent. That means individual creators cannot rely on platform safety alone — they must combine smart evidence preservation, targeted reporting, legal pressure, and clear public messaging.

Quick checklist (printable) — what to do in the first 24, 48, and 72 hours

First 0–24 hours (priority: preserve & report)

Don’t engage with anyone posting or sharing the content — engagement often amplifies reach.
Preserve the content: take multiple screenshots, download video files, save the original URL(s) and capture timestamps.
Export metadata: gather file headers, EXIF (if any), and export a HAR (HTTP Archive) from your browser to capture network requests.
Hash files for chain-of-custody (SHA-256 recommended).
Report to platform using the platform’s abuse or non-consensual nudity policy; attach evidence and request immediate removal and preservation of logs.
Contact your lawyer or an attorney experienced in image-based sexual abuse (IBSA) or privacy law — even an initial call helps plan preservation letters and subpoenas.

24–72 hours (priority: escalation & comms)

Send takedown requests to host, registrar, and CDN where applicable (see templates below).
Send a preservation letter asking the platform or host to retain all records; ask your lawyer to follow up with a subpoena readiness plan.
Notify law enforcement and file a report for image-based sexual abuse or harassment where local laws allow.
Prepare public messaging and a short FAQ for your audience; share responsibly to minimize spread (examples below).
Set up monitoring (Google Alerts, reverse-image search, and paid monitoring if budget allows). See tools and monitoring approaches in our micro tools roundup.

First week (priority: stabilization & legal planning)

Follow up with platforms and hosts if content hasn’t been removed within their stated response windows.
Consider civil remedies (tort claims, statutory remedies, injunctions) based on jurisdiction; discuss strategy with counsel.
Start reputation repair: pin an official update, update bios, and reinforce community guidelines for fans.

Detailed evidence preservation steps (how to collect things correctly)

Evidence is the foundation of takedowns, legal cases, and law enforcement reports. Collect multiple, independent records.

1. Screenshots + full-page saves

Capture the post in desktop and mobile views so timestamps and UI elements are visible.
Use full-page save tools (SingleFile or Save Page WE) to create an HTML snapshot with embedded assets.

2. Download original media

Download videos and images instead of relying solely on screenshots. For videos, use yt-dlp or browser devtools to fetch the file.
Record the URL exactly and copy surrounding text, comments, and the poster's username/ID.

3. Export browser network logs (HAR) and console

Open Developer Tools > Network > Start recording > reload the page > Save as HAR. This preserves requests to CDNs and hosting origins. See edge and CDN patterns for why origin traces matter.
HAR files are valuable to technical teams and law enforcement because they show where assets originated.

4. Hash and timestamp

Compute a SHA-256 (or SHA-1) hash for each downloaded file. Command: sha256sum filename.jpg (or use a Mac/Windows equivalent).
Record UTC timestamps when each item was captured, and keep an evidence log (who captured it and how).

5. Use reverse image search & provenance tools

Run images through Google Reverse Image Search, TinEye, and Yandex to track reposts and timelines.
Tools like FotoForensics or InVID (see our review of deepfake detection tools) can show signs of manipulation and help substantiate claims of AI generation.

6. Preserve account details

Save profile pages of posters (username, profile ID, follower count), direct messages, and timestamps of interactions.
If the content appears on anonymous platforms, capture any metadata that might identify the uploader (e.g., IP in HAR or server logs).

How to report — platform-by-platform playbook (short)

Each platform has different reporting channels. When non-consensual sexual content is involved, use the platform’s sexual exploitation or non-consensual nudity category and request urgent review.

X/Threads/Meta: Use the non-consensual nudity or harassment reporting flow and attach your evidence. Include a request for account data preservation.
Instagram/TikTok: Use in-app reporting and the web forms. Escalate via Creator Support if you have a partner/creator account.
YouTube: Use 'Sexual content involving minors' or 'Non-consensual porn' report forms; request immediate takedown and preservation.
Reddit/4chan/Anon boards: Report to moderators and the site admins; capture the thread ID and permalink.
Hosting providers and registrars: Find abuse contacts via WHOIS and domain due-diligence and send takedown and preservation requests.

Who else to notify

Local law enforcement — file an image-based sexual abuse report if your jurisdiction allows.
National hotlines and reporting centres (e.g., in the UK or EU countries, online-safety regulators).
Your attorney — for preservation letters, subpoena planning, and civil remedies.
Trusted managers, agents, or platform partners — to coordinate comms and evidence collection.

Legal options and what each does

Legal remedies differ by country, but common options include:

Preservation letters to force platforms/hosts to retain logs and content.
DMCA takedowns if you hold copyright in the original image (often effective in the U.S.).
Civil claims for privacy invasion, defamation, or intentional infliction of emotional distress.
Criminal complaints under image-based sexual abuse laws or harassment statutes (where applicable).
Data protection complaints (GDPR) and “right to be forgotten” petitions in the EU — can force delisting and removals.

Note: In early 2026 regulators are more aggressive: state attorneys general and privacy regulators are treating systemic failures (for example, AI tools that allow obvious non-consensual outputs) as subject to investigation and potential enforcement. Use that context when escalating.

Practical takedown & legal templates

Platform takedown / abuse report (short)

Subject: Urgent — Non-consensual sexualised content of me (request removal & preservation)
I am [Full Name / Account Name], the person depicted in the content at: [URL]. This content is non-consensual and sexually explicit. I request immediate removal and preservation of all related data (timestamps, IP logs, account details). Evidence attached: screenshots, downloaded file(s) (SHA256: [hash]), HAR file. Please confirm removal and preservation within 24 hours and provide case ID. Thank you.

Preservation / preservation + subpoena readiness (for lawyers)

[Date]
To: Legal Abuse/Compliance Team
Regarding: preservation of evidence relating to non-consensual sexually explicit material depicting [Name / account]
Please preserve all content, including but not limited to: posts, files, direct messages, deleted items, IP logs, timestamps, and account registration data for the content accessible at [URL]. Preserve copies in native format and preserve backups for potential subpoena. Kindly confirm preservation and estimated retention period.

DMCA takedown template (if you own copyright)

To the Designated Agent,
I, [Name], am the copyright owner of the content identified below. I have a good faith belief that the use of the copyrighted material identified below is not authorized by the copyright owner, its agent, or the law. Description of copyrighted work: [original image]. Location of infringing material: [URL]. I hereby request removal under 17 U.S.C. § 512(c). My contact information: [address, email, phone]. I declare under penalty of perjury that the information in this notice is accurate. [Signature]

Crisis communications — short templates and guidance

Your response should be timely, factual, and minimize amplification of the content. Aim to inform your audience while directing them away from sharing the harmful material.

Short public statement (for pinned post or website)

I am aware of non-consensual manipulated images/videos of me circulating online. These are AI-generated and deeply harmful. I have reported and requested takedowns and am working with legal counsel and platforms. Please do not share the content. If you see it, report it and share this official statement instead. Thank you for your support.

Direct message to followers/supporters

Hey — I want to share that AI-generated sexual content of me has been posted without consent. I’m taking steps to remove it and pursue legal remedies. Please don’t reshare the material and report any posts you see. If you have screenshots or URLs, DM them to me directly (do not repost).

Technical escalation: contacting hosts, registrars, and CDNs

If removal on the surface platform is slow, take the content to the origin:

Use WHOIS and domain due-diligence to find hosting and registrar abuse contacts.
Send your takedown and preservation requests to abuse@[domain], legal@, and the registrar’s abuse contact.
Look up CDN providers (Cloudflare, Fastly, Akamai) and send them the same info — they can sometimes disable access quickly if content violates their policies. See edge-first patterns for why origin/CDN traces matter.

Monitoring and prevention — what to change now

Enable two-factor authentication on all accounts and rotate passwords.
Watermark key images and use lower-resolution profile images until the situation stabilises.
Disable auto-sync of cloud albums from devices connected to public sharing platforms.
Limit audience for sensitive posts and remove geotags and identifying metadata from future uploads.
Set up continuous monitoring: reverse-image alerts, paid monitoring services, and alerts for your name/handles. See our review of detection and monitoring tools for recommendations: deepfake detection tools.

Long-term remedies and advocacy

Beyond the immediate crisis, creators should:

Work with legal counsel on cease-and-desist, civil claims, and securing injunctions where necessary.
Document incidents to support policy changes with platforms and regulators. Collective evidence can influence enforcement and AI safety rules.
Consider membership in coalitions that lobby for better creator protections from AI misuse; 2026 has seen stronger regulator attention — use that momentum.

What to expect from platforms and regulators in 2026

Expect faster—but inconsistent—responses. Many platforms updated policies in late 2025 and early 2026 after high-profile failures, yet enforcement still varies by platform and region. Regulators (including state attorneys general in the U.S. and national data regulators in the EU) are more willing to open investigations into systemic platform failures. Cite these investigations when you escalate with a platform or host to add urgency. See recent coverage on regulatory shifts: Ofcom and privacy updates (2026).

Sample evidence checklist (copy this into your notes)

Exact URL(s) and permalink(s) of offending posts
Screenshots (desktop & mobile) with timestamps
Downloaded original media files (video & images)
SHA-256 hashes of each file
HAR files from network capture
Reverse-image search results (links)
Usernames and profile IDs of posters and sharers
Direct messages related to the post (if any)
Copy of your public statement and log of reports filed (case IDs)

Do’s & Don’ts (critical)

Do preserve everything and ask for preservation from platforms.
Do report quickly and follow up in writing.
Do get counsel early and use a preservation letter if needed.
Don’t repost or circulate the harmful material — that amplifies it and can hurt legal claims.
Don’t negotiate with or pay extortionists; contact law enforcement immediately.

Final notes

Non-consensual AI sexualisation is a fast-moving problem that intersects technology, law, and human rights. Acting quickly, documenting carefully, and coordinating legal and platform escalation are your best tools to stop harm and regain control.

Actionable takeaway — your 10-item immediate checklist

Do not engage with the content or users posting it.
Take desktop and mobile screenshots with timestamps.
Download every media file and compute SHA-256 hashes.
Export a HAR from your browser and save console/network logs.
Report to the platform(s) using non-consensual content categories and attach evidence.
Send preservation requests to platform and host; get confirmation numbers.
Contact an attorney experienced in image-based sexual abuse/privacy.
File a law enforcement report if applicable in your jurisdiction.
Publish a short official statement asking followers not to share the material.
Set up ongoing monitoring and change passwords/2FA immediately.

Call to action

If you’re a creator dealing with non-consensual AI sexualisation right now, save this article and use the templates above. Start by preserving evidence immediately and file reports with platforms and law enforcement. If you want a ready-to-use, downloadable checklist and editable email templates for takedowns and preservation letters, download the printable checklist and templates (or contact counsel) — and share this guide with other creators: quick, coordinated action reduces harm.

Unknown

Contributor

Senior editor and content strategist. Writing about technology, design, and the future of digital media. Follow along for deep dives into the industry's moving parts.