Why synthetic media needs a structured legal response
A voice recording that sounds like a client or a video that appears to show an endorsement can spread before anyone verifies its origin. The immediate problem is factual: what was published, by whom, to which audience, and with what consequence?
MeitY published FAQs on the February 2026 amendments to the IT Rules addressing synthetically generated information. The official explanation states that the amendments took effect on 20 February 2026. That makes platform duties and synthetic-media definitions a current research issue, but the relevant text and any later changes must still be checked for the particular dispute.
Preserve what the client actually encountered
Capture the exact URL, account handle, upload time if available, date of discovery and surrounding page context. Keep the original file where it can lawfully be obtained, along with the messages or advertisements that led the client to it. Record who collected each item.
A cropped screenshot may illustrate the harm while losing the information needed to identify the source. Preserve a working copy separately from the collected material. Avoid repeatedly forwarding intimate or sensitive content; restrict access and document the lawful purpose of handling it.
Identify the harm before selecting the legal route
Separate impersonation, financial fraud, reputational injury, privacy intrusion and unauthorised commercial endorsement. One synthetic recording may raise several issues, but each requires its own factual and legal foundation. The label deepfake alone does not establish every element of a claim.
Ask whether money was requested, whether the person was identifiable, whether a statement was presented as authentic, and whether there is evidence of publication or reliance. These questions help counsel select the relevant civil, criminal or platform process without assuming that every synthetic work is unlawful.
Prepare a precise platform complaint
Identify the complained-of items individually. Explain the impersonation or other harm in plain language, attach the necessary supporting material and specify the action requested. Keep the acknowledgement, reference number, submission time and subsequent replies.
Check the applicable intermediary category, grievance route and current rule before advising on a response deadline. The duties of a content creator, a hosting service and a significant social media intermediary need separate analysis. An official FAQ explains the framework; it should be read alongside the notified rules and relevant orders.
Frame urgent court research around the requested relief
If court intervention is being considered, build the research question around the facts and relief: an identifiable claimant, specified content, continuing circulation, demonstrated harm and a proposed order. Identify the proper forum and parties before drafting a wide request.
Research preservation and identification issues separately from removal or restraint. Check territorial connections, evidence needed for urgency, the scope of any proposed order and applicable speech protections. Do not assume that an order in another claimant's case automatically grants the same relief here.
Keep authenticity and attribution separate
An expert may help assess whether a recording has been manipulated. That assessment does not necessarily establish who created it, who uploaded it or who controlled a particular account. Preserve the distinction in both the evidence note and the draft pleadings.
Build a chronology that separates verified facts, client instructions and questions awaiting technical analysis. A carefully framed factual record gives both the forensic specialist and the legal researcher a clearer task.
Sources and further reading
Source check: 23 October 2026. Verify later amendments, notifications and judgments before applying this guide to a live matter.
Continue with preparing electronic evidence under the BSA.