AI deepfake abuse against women in India: what to do when an image is fake but the harm is real
A practical response to sexualised or humiliating AI images, audio or video: protect immediate safety, preserve evidence, report strategically and keep control of your choices.
In this guide
What deepfake abuse is
A deepfake is manipulated media that makes it appear a person said or did something they did not. When it is made or shared without consent to sexualise, shame, threaten or silence someone, the abuse is real even when the media is fake.
The person did not consent
Consent to a photograph, relationship or private message is not consent to editing or publishing sexualised media. Do not ask the survivor to prove that the original image was private before offering support.
The harm can move offline
Deepfakes can trigger threats, stalking, school or workplace consequences, family violence and so-called honour-based retaliation. Treat exposure, doxxing or threats as safety concerns, not only reputation problems.
Sharing is participation
Forwarding a fake image, downloading it for curiosity or naming the person in a group chat increases reach and harm. Save only what is needed for a report and do not circulate it.
You do not have to investigate alone
You can ask a trusted person, advocate or lawyer to help assess options. The survivor decides what support is safe unless an immediate safeguarding duty applies.
The first 24 hours
Prioritise safety and containment. There is no single correct order, and a survivor may need to pause before reporting.
Check the immediate risk
If someone is threatening violence, blackmail, forced marriage, self-harm or a visit to your home, move to a safer place, contact a trusted person and use emergency services where needed. Do not meet the creator alone.
Reduce further exposure
Use platform privacy controls, block accounts and ask trusted contacts not to tag, reply to or repost the material. Change passwords from a safer device if an account may be compromised.
Do not bargain with a blackmailer
Payment or sending more images rarely guarantees deletion. Keep the messages, avoid promising anything and seek a safety plan before responding. A threat is the abuser’s action, not the survivor’s fault.
Choose a small support team
Tell one or two people who will keep information confidential, respect your decisions and help with transport, work, school, housing or a report. Do not add relatives who may increase danger.
Care for the body and mind
Shock, shame, sleeplessness and panic are common responses to violation. Basic care and a trauma-informed counsellor can help; urgent medical or mental-health support is appropriate when needed.
Preserve evidence without spreading it
Evidence can support a platform, police or court response, but collecting it should not expose you to more harm.
Capture the context
Record the account name, URL, date and time, caption, comments, threats and where the content appeared. A screenshot that shows context can be more useful than a cropped image.
Keep an incident log
Write a short timeline of discovery, reports, contacts, threats, offline effects and any new accounts. Store it in a private account or device that the abuser cannot access.
Preserve originals safely
Do not edit, annotate or repeatedly download intimate material. If a service asks for a file, use its secure upload route and note what you sent. Ask an advocate about lawful evidence handling for a minor.
Request a case number
When you report, save confirmation emails, ticket numbers and the exact policy category selected. A follow-up is easier when each report has a date and reference.
Protect other people’s privacy
Blur or avoid unrelated faces, phone numbers and children in copies used for teaching or support. Evidence should not become another form of public circulation.
Reporting and legal options in India
Online abuse can involve several laws and reporting routes. Rules and platform procedures change, so use current official guidance and qualified legal help for a case-specific decision.
Report the post and the account
Choose the closest category such as non-consensual intimate imagery, impersonation, sexual exploitation, harassment or threats. Ask the platform to preserve records and stop re-uploads where its process allows.
Use India’s cybercrime route
The National Cyber Crime Reporting Portal can receive complaints about women and children and other cyber offences. Keep the acknowledgement and attach only what the form requests.
Ask about applicable offences
The Information Technology Act, Bharatiya Nyaya Sanhita and other laws may be relevant depending on the conduct, age of the person, threats and publication. A lawyer or legal-aid service can map facts to current provisions.
Protect a child’s identity
If the person depicted may be under 18, do not download or forward the material. Use child-safety reporting and seek immediate specialist advice; the priority is protection and lawful handling.
Record retaliation
If a school, employer, family member or platform punishes the survivor for the abuse, document the decision and ask for a written reason. Retaliation can require a separate support or complaint route.
For supporters, platforms and institutions
The person targeted should not have to carry the entire investigation, takedown effort and social fallout.
Supporters should ask, not seize control
Say: ‘I believe you. What would make today safer?’ Offer practical help, keep the material private and do not contact the creator or publish a warning without consent.
Schools and employers should contain harm
Provide a private contact, stop gossip and retaliation, preserve relevant records and make reasonable adjustments while a report is considered. Do not demand that the survivor repeatedly display the content.
Platforms should make reports usable
Reporting should be visible, accessible, fast enough for viral harm and clear about what happens next. Systems should detect re-uploads, protect reporters from retaliation and provide human review for high-risk cases.
Public discussion should centre consent
Media and leaders should avoid naming, linking or describing the image in ways that increase reach. Explain the perpetrator’s conduct and the system’s duty instead of asking what the woman did to invite it.
Keep agency at every step
The survivor may choose takedown, a police complaint, a private support plan, several routes or no immediate action. Support does not require surrendering decisions.
Questions people ask
Does a fake image count as abuse?
Yes. The violation is creating or sharing sexualised or humiliating content without consent; its falsity does not cancel the harm.
Should I send the image to prove it is fake?
No. Send it only through a trusted reporting or legal channel that asks for it. A link, account, timestamp and description may be enough for an initial report.
What if my family may hurt me?
Plan for safety before disclosure. Choose a safe device and person, consider housing and transport, and contact a local women’s or legal-aid service. A public confrontation can increase risk.
Can the internet remove everything?
Removal can be difficult and copies may persist. A focused record, repeated platform reports and support can still limit spread and create accountability; none of this is the survivor’s responsibility alone.
Related practical guides
Related issue guides
Sources and publication record
Draft prepared 14 September 2026; project-team editorial review pending · Sources checked .
- When justice fails: Why women can't get protection from AI deepfake abuseUN Women
- Information Technology Act, 2000India Code, Legislative Department
- Bharatiya Nyaya Sanhita, 2023India Code, Legislative Department
- National Cyber Crime Reporting PortalIndian Cybercrime Coordination Centre, Ministry of Home Affairs
- Digital Personal Data Protection Act, 2023India Code, Legislative Department