Desifakes | Ai Generated

: The Information Technology (IT) Rules mandate that social media intermediaries remove non-consensual deepfakes within 24 hours of a complaint. Provisions under the Bharatiya Nyaya Sanhita (BNS) also penalize forgery, defamation, and impersonation via digital means.

“That’s insane,” Riya whispered. “You didn’t stop her.”

Using brief audio samples to generate highly convincing, entirely fabricated speeches or voice notes in regional languages like Hindi, Urdu, Bengali, Tamil, or Punjabi.

Historically, video evidence was considered the ultimate truth. The psychological transition to understanding that "seeing is no longer believing" lags far behind the technology's adoption rate. desifakes ai generated

The Hyderabad-based fact-checking initiative Factly has been at the forefront, utilizing multiple AI detection tools to verify videos. Rakesh Dubbudu, its founder, points to tell-tale cues like unnatural body language, added sound, and the "brazenness" of what the video is trying to portray. However, he notes a growing challenge: "If the duration of a video clip is less than 10 seconds, which we see happening more often now, existing detectors can't conclusively prove that it is synthetically generated,".

The rapid evolution of artificial intelligence has democratized media creation, bringing both incredible creative tools and unprecedented ethical challenges. Among the most concerning developments in the digital landscape is the surge in AI-generated "desifakes."

In the same hour, a temple bell rings in Varanasi, the azan echoes in Old Delhi, a hymn rises from a church in Goa, and a farmer in Punjab thanks the morning sun. Not as competition — but as rhythm. : The Information Technology (IT) Rules mandate that

The vast majority of these AI-generated images are malicious, focusing on creating explicit, pornographic, or misleading content targeting women.

This misuse of AI has significant financial implications for the average person. A 2025 McAfee report revealed a startling statistic: , with victims losing an average of ₹34,500 to such scams. Even more concerning, 60% have seen AI-generated or deepfake content from influencers and online personalities, not just mainstream celebrities, indicating how rapidly this deceptive content is proliferating. The report named Shah Rukh Khan as the most exploited celebrity, followed by Alia Bhatt, used to promote fake giveaways, phishing links, or malicious downloads.

Filmmakers use AI to make aging superstars look decades younger or to respectfully recreate deceased icons for cameo appearances. “You didn’t stop her

On the enterprise side, platforms like VAR India launched "Faceoff," a multimodal AI platform powered by a proprietary Adaptive Cognito Engine. This "Make in India" solution analyzes eight biometric and behavioral cues—including facial micro-expressions, voice sentiment, and eye movement—delivering trust scores within seconds. Meanwhile, the government's IndiaAI initiative is scaling up Safe AI efforts to advance real-time deepfake detection and forensic analysis, ensuring AI systems deployed in India are reliable and secure.

The "Desi" prefix is crucial here. It denotes a specific targeting based on ethnic and regional identity. While global deepfake porn heavily features Western women, the infrastructure for Desi Fakes operates in the shadows of the internet—on encrypted Telegram channels, closed Discord servers, and localized dark web forums. These spaces operate on an economy of exchange: users trade "real" leaked images (a longstanding issue in South Asia, exacerbated by the proliferation of cheap smartphones) for "faked" AI-generated content, creating an endless feedback loop of digital sexual exploitation.

Beyond entertainment, the most dangerous application of DesiAI lies in the political arena. India is the world's largest democracy, and as elections approach, the digital space becomes a battlefield where truth is the first casualty.

Governments across South Asia are updating their legal frameworks to address synthetic media. Under various regional information technology acts, distributing non-consensual deepfakes or using AI to impersonate individuals maliciously can result in steep fines and imprisonment. However, enforcing these laws across decentralized, anonymous online spaces remains a significant hurdle. Media Literacy

To examine "Desi Fakes" is not merely to look at a technological aberration, but to peer into a dark nexus of post-colonial desire, patriarchal entitlement, cyber-misogyny, and the unique socio-cultural vulnerabilities of the Subcontinent. It is a crisis that takes a global technology and weaponizes it through deeply local pathologies.