Synopsis: Deepfakes are becoming a huge threat. This AI-based technology has been used to incite violence and disrepute people. Recently, it was used allegedly used to incite violence in the US.
What are Deepfakes?
- Deepfakes are modified images, text, audio, and video or synthetic media, created with the help of Artificial Intelligence.
- It generates a fake version from an original or real audio-visual content by superimposing new audio or image on an existing media file.
- For example; with the use of AI, the face of a person in an original video can be replaced with the face of another person (Morphing). Now the modified face will mimic the head movements, vocal patterns, and facial expressions of the original one.
- Media is manipulated with such sharpness that it becomes almost impossible to identify the difference between fake and real media. It can only be identified by using AI tools.
Threats of Deep Fakes
- Deep fakes can be used to disrupt the democratic processes like elections in any country.
- It is believed that Capitol Hill violence was incited by using deep fake media which caused misinformation and disinformation.
- Deep Fakes are used to stain the reputations of individuals and spread propaganda against them.
- According to a deep fake tracking research organization, in the month of October alone, over 100,000 fake nude images of women were circulating online.
- Real images for that purpose were acquired through social media accounts.
- The existence of deep fakes causes that much distrust among the public that any true evidence of a crime can easily be dismissed as fake.
- These technologies can be used by terrorist organizations and insurgents to further their agenda of destabilizing state governments. They can spread false information about institutions, public policy, and politicians for this purpose.
- The existing legal framework of many countries including India does not criminalize deep fakes.
Regulations of Deep fakes
In US: As per US law, Social Media Companies cannot be held responsible for the posts on their platforms.
In India: Some provisions in the Indian Penal Code (IPC) and Information Technology Act, 2000 criminalize a few related cybercrimes. But there is no specific law as of now to deal with deep fakes.
Issues in India’s legal system
- Sections 66E, 67, 67A, and 72 of the IT act deal with the violation of privacy and publishing or transmitting obscene material in electronic form.
- But these provisions can also result in penal consequences for the victim, for voluntarily producing such material.
- The Information Technology Intermediary Guidelines (Amendment) Rules, 2018 are insufficient to tackle content manipulation on digital platforms.
- Guidelines require that intermediate companies take necessary steps for the removal of illegal content.
- During the 2019 general election of India, the election commission gave out instructions on the use of social media for election campaigns. Social media companies also agreed to take action to prevent any violations.
- However, it has been reported that social media platforms like WhatsApp were used for spreading misinformation and propaganda during the election.
What are the steps to be taken?
Existing laws are not enough to protect individuals against deep fakes. Only AI-generated tools can be effective in detection.
- AI-based detection tools with the capability to detect deep fakes must be invented as soon as possible.
- In 2020, the University of Washington and Microsoft arranged a workshop with experts to discuss how to avoid deep fake technology from harmfully affecting the 2020 U.S. presidential election. The workshop concluded with the following suggestions:
- Deep Fakes must be included under hateful manipulated media, propaganda, and disinformation campaigns.
- Journalists should be provided with tools to examine the authenticity of images, video, and audio recordings. For that, they require proper training and resources.
- Policymakers need to understand how deep fakes can threaten polity, society, economy, culture, individuals, and communities.
The best way to deal with this menace is AI-backed technological tools to detect and prevent deep fakes. These tools must be invented by the countries in cooperation as soon as possible. If steps are not taken immediately, these technologies are even capable of invoking wars among countries, in this information age.