Daily Editorial Analysis

Editorial Analysis based on “Deepfakes in elections: They have shaken our faith in our own judgement” which was published in The Hindu.

Introduction:

• Deepfakes represent a new and sophisticated form of manipulation in the political sphere, causing considerable anxiety.
• Despite a history of misinformation and manipulation in politics, deepfakes present a unique challenge due to their ability to deceive even vigilant observers.

Origin:

• The term "deepfake" is derived from a combination of "deep learning" and "fake."
• The technology relies on deep learning techniques, particularly generative adversarial networks (GANs), first introduced by Ian Goodfellow and his team in 2014.
• GANs consist of two neural networks, a generator, and a discriminator, that work together to create synthetic data that closely resembles accurate data. This breakthrough laid the foundation for the development of deepfake technology.

Impacts of Deepfakes:

• Impact on trust in Information and Media Institutions:
  • Historically, there has been a level of trust in our ability to convey truth from falsehood, aided by alternative sources, media institutions, and regulatory bodies.
  • While misinformation existed, there was confidence in our collective ability to verify and evaluate information.
  • Deepfakes undermine this trust by introducing doubt about the authenticity of what we see and hear, eroding our confidence in our own judgement.
• Impact of Social Media on Information Consumption:
  • Social media has become the primary platform for information consumption, leading to two significant shifts:
    • Context Collapse: The collapse of contextual cues, such as source credibility or expertise, undermines our ability to assess the reliability of information.
    • Information Overload: The overwhelming volume of information on social media makes it difficult to engage thoughtfully, leading to reliance on intuition and fragmented impressions.
• Challenges Posed by Context Collapse:
  • In the age of social media, information is divorced from its original context, making it harder to discern trustworthy sources.
  • We consume information through the same interface regardless of its source or credibility thus it becomes difficult to differentiate between reliable and unreliable sources.
• Impact of Information Overload:
  • Information overload affects our ability to critically evaluate information, as we rely on quick judgments and intuition.
  • Algorithms play a significant role in shaping our media consumption, making information based on profit and engagement rather than accuracy..
• Need for Clarity and Solutions:
  • Clarity between what is real, fake, and "really fake" is essential for managing the spread of deepfakes effectively.
  • Technological solutions for detecting deepfakes must be complemented by educational initiatives to promote media literacy and critical thinking skills.
  • Transparency and accountability in digital platforms can help mitigate the effects of information overload and restore trust in online content.
• Privacy concerns:
  • Deepfakes raise concerns regarding privacy and consent, as individuals' identities can be easily manipulated without their knowledge or permission.
• Security Implications:
  • The rise of deepfake technology presents significant security challenges. For instance, the technology can deceive facial recognition systems, bypassing security measures and compromising sensitive information.
  • Deepfakes can also impersonate individuals, creating opportunities for identity theft or social engineering attacks. This poses a substantial risk to individuals, organizations, and even national security.

Government initiatives:

• Responsibility for Social Media Platforms: The IT Act and IT Rules say that social media platforms must quickly delete fake videos or photos. If they don't, they could face up to three years in jail or a fine of Rs 1 lakh.
• Impersonation Penalties: According to Section 66D of the IT Act, people who trick others by pretending to be someone else online could go to jail for up to three years and be fined up to one lakh rupees.
• Preventing Impersonation: Social media platforms have to make sure users don't post content pretending to be someone else, as per Rule 3(1)(b)(vii).
• Quick Removal of Impersonation Content: If someone complains about content pretending to be someone else, social media platforms must remove it within 24 hours, according to Rule 3(2)(b).
• Fact Check Unit: The Fact Check Unit, set up under the IT rules in November 2019, aims to stop fake news and misinformation. It also lets people report suspicious information about the Government of India easily.

Conclusion:

• Deepfakes represent a significant challenge to our ability to discern truth from falsehood in the digital age.
• Addressing the underlying issues of context collapse and information overload is crucial for restoring trust in our ability to evaluate information critically.
• A multi-stakeholder approach involving governments, tech companies, media organizations, and civil society is necessary to mitigate the impact of deepfakes on democratic processes.

Download PDF