#FactCheck-AI-Generated Video Falsely Linked to Fresh US Strikes on Iran

The World Economic Forum reported that AI-generated misinformation and disinformation are the second most likely threat to present a material crisis on a global scale in 2024 at 53% (Sept. 2023). Artificial intelligence is automating the creation of fake news at a rate disproportionate to its fact-checking. It is spurring an explosion of web content mimicking factual articles that instead disseminate false information about grave themes such as elections, wars and natural disasters. 

According to a report by the Centre for the Study of Democratic Institutions, a Canadian think tank, the most prevalent effect of Generative AI is the ability to flood the information ecosystem with misleading and factually-incorrect content. As reported by Democracy Reporting International during the 2024 elections of the European Union, Google's Gemini, OpenAI’s ChatGPT 3.5 and 4.0, and Microsoft’s AI interface ‘CoPilot’ were inaccurate one-third of the time when engaged for any queries regarding the election data.   Therefore, a need for an innovative regulatory approach like regulatory sandboxes which can address these challenges while encouraging responsible AI innovation is desired.

What Is AI-driven Misinformation?

False or misleading information created, amplified, or spread using artificial intelligence technologies is AI-driven misinformation. Machine learning models are leveraged to automate and scale the creation of false and deceptive content. Some examples are deep fakes, AI-generated news articles, and bots that amplify false narratives on social media.

The biggest challenge is in the detection and management of AI-driven misinformation. It is difficult to distinguish AI-generated content from authentic content, especially as these technologies advance rapidly. 

AI-driven misinformation can influence elections, public health, and social stability by spreading false or misleading information. While public adoption of the technology has undoubtedly been rapid, it is yet to achieve true acceptance and actually fulfill its potential in a positive manner because there is widespread cynicism about the technology - and rightly so. The general public sentiment about AI is laced with concern and doubt regarding the technology’s trustworthiness, mainly due to the absence of a regulatory framework maturing on par with the technological development. 

Regulatory Sandboxes: An Overview

Regulatory sandboxes refer to regulatory tools that allow businesses to test and experiment with innovative products, services or businesses under the supervision of a regulator for a limited period. They engage by creating a controlled environment where regulators allow businesses to test new technologies or business models with relaxed regulations.

Regulatory sandboxes have been in use for many industries and the most recent example is their use in sectors like fintech, such as the UK’s Financial Conduct Authority sandbox. These models have been known to encourage innovation while allowing regulators to understand emerging risks. Lessons from the fintech sector show that the benefits of regulatory sandboxes include facilitating firm financing and market entry and increasing speed-to-market by reducing administrative and transaction costs. For regulators, testing in sandboxes informs policy-making and regulatory processes. Looking at the success in the fintech industry, regulatory sandboxes could be adapted to AI, particularly for overseeing technologies that have the potential to generate or spread misinformation.

The Role of Regulatory Sandboxes in Addressing AI Misinformation

Regulatory sandboxes can be used to test AI tools designed to identify or flag misinformation without the risks associated with immediate, wide-scale implementation. Stakeholders like AI developers, social media platforms, and regulators work in collaboration within the sandbox to refine the detection algorithms and evaluate their effectiveness as content moderation tools.

These sandboxes can help balance the need for innovation in AI and the necessity of protecting the public from harmful misinformation. They allow the creation of a flexible and adaptive framework capable of evolving with technological advancements and fostering transparency between AI developers and regulators. This would lead to more informed policymaking and building public trust in AI applications.

CyberPeace Policy Recommendations 

Regulatory sandboxes offer a mechanism to predict solutions that will help to regulate the misinformation that AI tech creates. Some policy recommendations are as follows: 

1. Create guidelines for a global standard for including regulatory sandboxes that can be adapted locally and are useful in ensuring consistency in tackling AI-driven misinformation.
2. Regulators can propose to offer incentives to companies that participate in sandboxes. This would encourage innovation in developing anti-misinformation tools, which could include tax breaks or grants.
3. Awareness campaigns can help in educating the public about the risks of AI-driven misinformation and the role of regulatory sandboxes can help manage public expectations.
4. Periodic and regular reviews and updates to the sandbox frameworks should be conducted to keep pace with advancements in AI technology and emerging forms of misinformation should be emphasized.

‍

Conclusion and the Challenges for Regulatory Frameworks

Regulatory sandboxes offer a promising pathway to counter the challenges that AI-driven misinformation poses while fostering innovation. By providing a controlled environment for testing new AI tools, these sandboxes can help refine technologies aimed at detecting and mitigating false information. This approach ensures that AI development aligns with societal needs and regulatory standards, fostering greater trust and transparency. With the right support and ongoing adaptations, regulatory sandboxes can become vital in countering the spread of AI-generated misinformation, paving the way for a more secure and informed digital ecosystem.

References

• https://www.thehindu.com/sci-tech/technology/on-the-importance-of-regulatory-sandboxes-in-artificial-intelligence/article68176084.ece
• https://www.oecd.org/en/publications/regulatory-sandboxes-in-artificial-intelligence_8f80a0e6-en.html
• https://www.weforum.org/publications/global-risks-report-2024/
• https://democracy-reporting.org/en/office/global/publications/chatbot-audit#Conclusions

‍

Introduction

The Ministry of Electronics and Information Technology (MeitY) issued an advisory on March 1 2024, urging platforms to prevent bias, discrimination, and threats to electoral integrity by using AI, generative AI, LLMs, or other algorithms. The advisory requires that AI models deemed unreliable or under-tested in India must obtain explicit government permission before deployment. While leveraging Artificial Intelligence models, Generative AI, software, or algorithms in their computer resources, Intermediaries and platforms need to ensure that they prevent bias, discrimination, and threats to electoral integrity. As Intermediaries are required to follow due diligence obligations outlined under “Information Technology (Intermediary Guidelines and Digital Media Ethics Code)Rules, 2021, updated as of 06.04.2023”. This advisory is issued to urge the intermediaries to abide by the IT rules and regulations and compliance therein.  

Key Highlights of the Advisories

• Intermediaries and platforms must ensure that users of Artificial Intelligence models/LLM/Generative AI, software, or algorithms do not allow users to host, display, upload, modify, publish, transmit, store, update, or share unlawful content, as per Rule 3(1)(b) of the IT Rules. 
• The government emphasises intermediaries and platforms to prevent bias or discrimination in their use of Artificial Intelligence models, LLMs, and Generative AI, software, or algorithms, ensuring they do not threaten the integrity of the electoral process.
• The government requires explicit permission to use deemed under-testing or unreliable AI models, LLMs, or algorithms on the Indian internet. Further, it must be deployed with proper labelling of potential fallibility or unreliability.  Further, users can be informed through a consent popup mechanism.
• The advisory specifies that all users should be well informed about the consequences of dealing with unlawful information on platforms, including disabling access, removing non-compliant information, suspension or termination of access or usage rights of the user to their user account and imposing punishment under applicable law. It entails that users are clearly informed, through terms of services and user agreements, about the consequences of engaging with unlawful information on the platform. 
• The advisory also indicates measures advocating to combat deepfakes or misinformation. The advisory necessitates identifying synthetically created content across various formats, advising platforms to employ labels, unique identifiers, or metadata to ensure transparency. Furthermore, the advisory mandates the disclosure of software details and tracing the first originator of such synthetically created content.

Rajeev Chandrasekhar, Union Minister of State for IT, specified that

“Advisory is aimed at the Significant platforms, and permission seeking from Meity is only for large platforms and will not apply to startups. Advisory is aimed at untested AI platforms from deploying on the Indian Internet. Process of seeking permission , labelling & consent based disclosure to user about untested platforms is insurance policy to platforms who can otherwise be sued by consumers. Safety & Trust of India's Internet is a shared and common goal for Govt, users and Platforms.”

Conclusion

MeitY's advisory sets the stage for a more regulated Al landscape. The Indian government requires explicit permission for the deployment of under-testing or unreliable Artificial Intelligence models on the Indian Internet. Alongside intermediaries, the advisory also applies to digital platforms that incorporate Al elements. Advisory is aimed at significant platforms and will not apply to startups. This move safeguards users and fosters innovation by promoting responsible AI practices, paving the way for a more secure and inclusive digital environment.

References 

• https://regmedia.co.uk/2024/03/04/meity_ai_advisory_1_march.pdf
• https://economictimes.indiatimes.com/tech/technology/govts-ai-advisory-will-not-apply-to-startups-mos-it-rajeev-chandrasekhar/articleshow/108197797.cms?from=mdr
• https://www.meity.gov.in/writereaddata/files/Advisory%2015March%202024.pdf

‍

What are Deepfakes? 

A deepfake is essentially a video of a person in which their face or body has been digitally altered so that they appear to be someone else, typically used maliciously or to spread false information. Deepfake technology is a method for manipulating videos, images, and audio utilising powerful computers and deep learning. It is used to generate fake news and commit financial fraud, among other wrongdoings. It overlays a digital composite over an already-existing video, picture, or audio; cybercriminals use Artificial Intelligence technology. The term deepfake was coined first time in 2017 by an anonymous Reddit user, who called himself deepfake. 

Deepfakes works on a combination of AI and ML, which makes the technology hard to detect by Web 2.0 applications, and it is almost impossible for a layman to see if an image or video is fake or has been created using deepfakes. In recent times, we have seen a wave of AI-driven tools which have impacted all industries and professions across the globe. Deepfakes are often created to spread misinformation. There lies a key difference between image morphing and deepfakes. Image morphing is primarily used for evading facial recognition, but deepfakes are created to spread misinformation and propaganda. 

Issues Pertaining to Deepfakes in India 

Deepfakes are a threat to any nation as the impact can be divesting in terms of monetary losses, social and cultural unrest, and actions against the sovereignty of India by anti-national elements. Deepfake detection is difficult but not impossible. The following threats/issues are seen to be originating out of deep fakes: 

• Misinformation: One of the biggest issues of Deepfake is misinformation, the same was seen during the Russia-Ukraine conflict, where in a deepfake of Ukraine’s president, Mr Zelensky, surfaced on the internet and caused mass confusion and propaganda-based misappropriation among the Ukrainians. 
• Instigation against the Union of India: Deepfake poses a massive threat to the integrity of the Union of India, as this is one of the easiest ways for anti-national elements to propagate violence or instigate people against the nation and its interests. As India grows, so do the possibilities of anti-national attacks against the nation. 
• Cyberbullying/ Harassment: Deepfakes can be used by bad actors to harass and bully people online in order to extort money from them. 
• Exposure to Illicit Content: Deepfakes can be easily used to create illicit content, and oftentimes, it is seen that it is being circulated on online gaming platforms where children engage the most. 
• Threat to Digital Privacy: Deepfakes are created by using existing videos. Hence, bad actors often use photos and videos from Social media accounts to create deepfakes, this directly poses a threat to the digital privacy of a netizen.  
• Lack of Grievance Redressal Mechanism: In the contemporary world, the majority of nations lack a concrete policy to address the aspects of deepfake. Hence, it is of paramount importance to establish legal and industry-based grievance redressal mechanisms for the victims. 
• Lack of Digital Literacy: Despite of high internet and technology penetration rates in India, digital literacy lags behind, this is a massive concern for the Indian netizens as it takes them far from understanding the tech, which results in the under-reporting of crimes. Large-scale awareness and sensitisation campaigns need to be undertaken in India to address misinformation and the influence of deepfakes. 

How to spot deepfakes? 

Deepfakes look like the original video at first look, but as we progress into the digital world, it is pertinent to establish identifying deepfakes in our digital routine and netiquettes in order to stay protected in the future and to address this issue before it is too late. The following aspects can be kept in mind while differentiating between a real video and a deepfake

• Look for facial expressions and irregularities: Whenever differentiating between an original video and deepfake, always look for changes in facial expressions and irregularities, it can be seen that the facial expressions, such as eye movement and a temporary twitch on the face, are all signs of a video being a deepfake. 
• Listen to the audio: The audio in deepfake also has variations as it is imposed on an existing video, so keep a check on the sound effects coming from a video in congruence with the actions or gestures in the video. 
• Pay attention to the background: The most easiest way to spot a deepfake is to pay attention to the background, in all deepfakes, you can spot irregularities in the background as, in most cases, its created using virtual effects so that all deepfakes will have an element of artificialness in the background. 
• Context and Content: Most of the instances of deepfake have been focused towards creating or spreading misinformation hence, the context and content of any video is an integral part of differentiating between an original video and deepfake. 
• Fact-Checking: As a basic cyber safety and digital hygiene protocol, one should always make sure to fact-check each and every piece of information they come across on social media. As a preventive measure, always make sure to fact-check any information or post sharing it with your known ones.  
• AI Tools: When in doubt, check it out, and never refrain from using Deepfake detection tools like- Sentinel, Intel’s real-time deepfake detector - Fake catcher, We Verify, and Microsoft’s Video Authenticator tool to analyze the videos and combating technology with technology. 

Recent Instance 

A deepfake video of actress Rashmika Mandanna recently went viral on social media, creating quite a stir. The video showed a woman entering an elevator who looked remarkably like Mandanna. However, it was later revealed that the woman in the video was not Mandanna, but rather, her face was superimposed using AI tools. Some social media users were deceived into believing that the woman was indeed Mandanna, while others identified it as an AI-generated deepfake. The original video was actually of a British-Indian girl named Zara Patel, who has a substantial following on Instagram. This incident sparked criticism from social media users towards those who created and shared the video merely for views, and there were calls for strict action against the uploaders. The rapid changes in the digital world pose a threat to personal privacy; hence, caution is advised when sharing personal items on social media​​.

Legal Remedies

Although Deepfake is not recognised by law in India, it is indirectly addressed by Sec. 66 E of the IT Act, which makes it illegal to capture, publish, or transmit someone's image in the media without that person's consent, thus violating their privacy. The maximum penalty for this violation is ₹2 lakh in fines or three years in prison. The DPDP Act's applicability in 2023 means that the creation of deepfakes will directly affect an individual's right to digital privacy and will also violate the IT guidelines under the Intermediary Guidelines, as platforms will be required to exercise caution while disseminating and publishing misinformation through deepfakes. The indirect provisions of the Indian Penal Code, which cover the sale and dissemination of derogatory publications, songs and actions, deception in the delivery of property, cheating and dishonestly influencing the delivery of property, and forgery with the intent to defame, are the only legal remedies available for deepfakes. Deep fakes must be recognized legally due to the growing power of misinformation. The Data Protection Board and the soon-to-be-established fact-checking body must recognize crimes related to deepfakes and provide an efficient system for filing complaints.

Conclusion 

Deepfake is an aftermath of the advancements of Web 3.0 and, hence is just the tip of the iceberg in terms of the issues/threats from emerging technologies. It is pertinent to upskill and educate the netizens about the keen aspects of deepfakes to stay safe in the future. At the same time, developing and developed nations need to create policies and laws to efficiently regulate deepfake and to set up redressal mechanisms for victims and industry. As we move ahead, it is pertinent to address the threats originating out of the emerging techs and, at the same time, create a robust resilience for the same. 

References 

• https://www.drishtiias.com/daily-updates/daily-news-analysis/deepfake-technology
• https://timesofindia.indiatimes.com/gadgets-news/explained-what-are-deepfakes-and-how-to-spot-one/articleshow/90916962.cms
• https://www.unite.ai/best-deepfake-detector-tools-and-techniques/