A disturbing deepfake video depicting popular Indian actress Rashmika Mandanna recently went viral, provoking widespread condemnation and underscoring the urgent need for legal safeguards.
The digitally altered video shows Mandanna’s face realistically superimposed onto another woman’s body in an elevator. Its online circulation outraged fans and highlighted the proliferation of deepfakes enabled by advanced AI.
What Exactly Are Deepfake Videos?
The name “deepfake” combines the terms “deep learning” and “fake.” Deep learning is a form of AI that can analyze and recreate patterns in data like images, audio, and video.
Deepfake technology uses deep learning algorithms to replace a person’s face or body in an original video with someone else’s. The resulting fake footage looks strikingly realistic and authentic.
How do I Spot Fake Videos?
While convincing, deepfakes do contain subtle clues that they’ve been manipulated:
- Unnatural eye blinking or movements
- Mismatched colors, lighting, or shadows
- Lip movements are not synchronized with audio.
- Strange body shapes, motions, or anatomy
- Artificial-looking facial movements
Reverse image searches and deepfake detection tools can also help uncover fakery. But deepfakes are becoming more sophisticated and difficult to detect.
Deepfakes Enable Revenge Porn and Abuse
The Rashmika Mandanna video represents a disturbing trend of using deepfake video technology for unethical purposes like revenge porn, harassment, and abuse.
Deepfakes typically target women, superimposing their images into pornographic videos without consent. Their proliferation enables new forms of sexual exploitation.
AI Advances Make Deepfakes Increasingly Realistic
Recent AI breakthroughs have massively improved deepfakes’ realism, making them harder to identify as fabrications and more dangerous as disinformation tools.
Apps even allow anyone to easily generate deep fakes on their phone for free. Critics fear deepfakes could become an epidemic with society-wide impacts.
Legal and Technological Solutions are Needed
In response to public figures like Mandanna being victimized, experts advocate both legal deterrents and technological defenses against deepfakes.
Proposed laws would make creating nonconsensual intimate deepfakes a criminal offense. Meanwhile, researchers are racing to enhance AI-based deepfake detection techniques.
Deepfake Detection is Still Limited
While promising detection tools are emerging, most remain imperfect and not widely adopted by non-experts. The onus still lies on users to carefully scrutinize questionable videos.
Until better solutions emerge, the public must exercise caution in assessing the authenticity of any provocative images or footage of public figures.
Outrage Over Mandanna Video Renews Calls for Action
The outrage over the Rashmika Mandanna deepfake underscores the need for concrete action from lawmakers, tech companies, and society as a whole.
If unchecked, the proliferation of deepfakes enabled by ever-improving AI threatens privacy, trust in information, and institutions. An urgent collective response is required to counter this complex technological menace.
