AI-generated fake voices and videos are raising hard questions for Indian elections: how to act quickly without giving officials or platforms a broad licence to silence legitimate speech. In a Statesman opinion piece, Siddharth Roy argues for clearer rules focused on demonstrable harm, alongside safeguards for political expression and better ways to check where digital content came from.
Watch Desk analysis
What happened
Roy points to the Election Commission of India’s response during the 2024 general election: parties were directed to avoid deepfakes and remove misleading content within three hours of being notified, and the Commission launched a public Myth versus Reality register. He argues that corrections can struggle to catch up with viral material, especially when a convincing cloned voice travels through private messaging groups.
The columnist says existing legal provisions may apply to some cases, including impersonation, identity theft, defamation and unlawful intimate imagery, but that no single provision automatically covers every deepfake. He proposes clearer definitions centred on harms such as fraudulent electoral messages, financial scams and deliberate fabrication intended to cause identifiable harm. Those are recommendations in the opinion piece, not a description of a new law.
Roy also proposes safeguards for people wrongly affected by takedowns, clearer explanations of moderation decisions and stronger scrutiny of official assessments. For checking media, he points to Content Credentials, which can record information about a file’s origin and editing. He notes an important catch: metadata can be removed, and a genuine recording can still be edited to mislead.
Why it matters
A false clip can spread quickly; a rushed removal can also suppress satire, reporting or criticism. That makes the choice more complicated than simply asking platforms to take down suspicious material faster. The rules need to account for who made or circulated a clip, the context, intent and evidence of harm, while leaving room to challenge a mistaken decision.
The article offers readers a useful distinction between detecting synthetic media and deciding what should happen to it. A provenance record may help establish where a file came from, but it cannot settle whether the content is truthful or fair.
Our read
Roy’s strongest point is that speed and fairness have to travel together. A quick correction may be valuable, but a system that treats every altered clip as criminal or every automated flag as proof risks turning election protection into a shortcut around scrutiny. Provenance tools can help, not pronounce judgement on our behalf. Democracy, inconveniently, still requires people to explain their decisions.
What to watch
- Whether India develops clearer rules for harmful synthetic media while protecting satire and political criticism.
- How the Election Commission explains evidence and uncertainty in its election-related assessments.
- Whether platforms give people a practical way to appeal content removals.
- Whether provenance tools become useful in practice, including when metadata is missing or misleading.
Discussion spark: In an election, should platforms be required to remove suspected deepfakes within hours, or should they wait for stronger evidence even if a false clip keeps spreading?
Sources and evidence
- Source update (27 October 2026, 02:53 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.