#Factcheck-False Claims of Houthi Attack on Israel’s Ashkelon Power Plant

Executive Summary:

A morphed video of the actor Anup Soni popular on social media promoting IPL betting Telegram channel is found to be fake. The audio in the morphed video is produced through AI voice cloning. AI manipulation was identified by AI detection tools and deepfake analysis tools. In the original footage Mr Soni explains a case of crime, a part of the popular show Crime Patrol which is unrelated to betting.  Therefore, it is important to draw the conclusion that Anup Soni is in no way associated with the betting channel.

‍

Claims:

The facebook post claims the IPL betting Telegram channel which belongs to Rohit Khattar is promoted by Actor Anup Soni.

Fact Check:

Upon receiving the post, the CyberPeace Research Team closely analyzed the video and found major discrepancies which are mostly seen in AI-manipulated videos. The lip sync of the video does not match the audio. Taking a cue from this we analyzed using a Deepfake detection tool by True Media. It is found that the voice of the video is 100% AI-generated. 

We then extracted the audio and checked in an audio Deepfake detection tool named Hive Moderation. Hive moderation found the audio to be 99.9% AI-Generated.

‍

We then divided the video into keyframes and reverse searched one of the keyframes and found the original video uploaded by the YouTube channel named LIV Crime.

Upon analyzing we found that in the 3:18 time frame the video was edited, and altered with an AI voice.

‍

Hence, the viral video is an AI manipulated video and it’s not real. We have previously debunked such AI voice manipulation with different celebrities and politicians to misrepresent the actual context. Netizens must be careful while believing in such AI manipulation videos.

Conclusion:

In conclusion, the viral video claiming that  IPL betting Telegram channel promotion by actor Anup Soni is false. The video has been manipulated using AI voice cloning technology, as confirmed by both the Hive Moderation AI detector and the True Media AI detection tool. Therefore, the claim is baseless and misleading.

• Claim:  An IPL betting Telegram channel belonging to Rohit Khattar promoted by Actor Anup Soni.
• Claimed on: Facebook
• Fact Check: Fake & Misleading

Introduction

In a setback to the Centre, the Bombay High Court on Friday 20th September 2024, struck down the provisions under IT Amendment Rules 2023, which empowered the Central Government to establish Fact Check Units (FCUs) to identify ‘fake and misleading’ information about its business on social media platforms. 

Chronological Overview

• On 6th April 2023, the Ministry of Electronics and Information Technology (MeitY) notified the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules, 2023 (IT Amendment Rules, 2023). These rules introduced new provisions to establish a fact-checking unit with respect to “any business of the central government”. This amendment was done In exercise of the powers conferred by section 87 of the Information Technology Act, 2000. (IT Act).
• On 20 March 2024, the Central Government notified the Press Information Bureau (PIB) as FCU under rule 3(1)(b)(v) of the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Amendment Rules 2023 (IT Amendment Rules 2023). 
• The next day on  21st March 2024, the Supreme Court stayed the Centre's decision on notifying PIB -FCU, considering the pendency of the proceedings before the High Court of Judicature at Bombay. A detailed analysis covered by CyberPeace on the Supreme Court Stay decision can be accessed here.
• In the latest development, the Bombay High Court on 20th September 2024, struck down the provisions under IT Amendment Rules 2023, which empowered the Central Government to establish Fact Check Units (FCUs) to identify ‘fake and misleading’ information about its business on social media platforms.  

Brief Overview of Bombay High Court decision dated 20th September 2024

Justice AS Chandurkar was appointed as the third judge after a split verdict in January 2023 by a division bench consisting of Justices Gautam Patel and Neela Gokhal.  As a Tie-breaker judge' Justice AS Chandurkar delivered the decision striking down provisions for setting up a Fact Check Unit under IT amendment 2023 rules. Striking down the Centre's proposed fact check unit provision, Justice A S Chandurkar of Bombay High Court also opined that there was no rationale to undertake an exercise in determining whether information related to the business of the Central govt was fake or false or misleading when in digital form but not doing the same when such information was in print. It was also contended that there is no justification to introduce an FCU only in relation to the business of the Central Government. Rule 3(1)(b)(v) has a serious chilling effect on the exercise of the freedom of speech and expression under Article 19(1)(a) of the Constitution since the communication of the view of the FCU will result in the intermediary simply pulling down the content for fear of consequences or losing the safe harbour provision given under IT Act.  

Justice Chandurkar held that the expressions ‘fake, false or misleading’ are ‘vague and overbroad’, and that the ‘test of proportionality’ is not satisfied. Rule 3(1)(b)(v), was violative of Articles 14 and 19 (1) (a) and 19 (1) (g) of the Constitution and it is “ultra vires”, or beyond the powers, of the IT Act.

Role of Expert Organisations in Curbing Mis/Disinformation and Fake News

In light of the recent developments, and the rising incidents of Mis/Disinformation and Fake News it becomes significantly important that we all stand together in the fight against these challenges. The actions against Mis/Disinformation and fake news should be strengthened by collective efforts, the expert organisations like CyberPeace Foundation plays an key role in enabling and encouraging netizens to exercise caution and rely on authenticated sources, rather than solely rely on govt FCU to block the content. 

Mis/Disinformation and Fake News should be stopped, identified and countered by netizens at the very first stage of its spread. In light of the Bombay High Court's decision to stuck down the provision related to setting up the FCU by the Central Government, it entails that the government's intention to address misinformation related solely to its business/operations may not have been effectively communicated in the eyes of the judiciary.

It is high time to exercise collective efforts against Mis/Disinformation and Fake News and support expert organizations who are actively engaged in conducting proactive measures, and campaigns to target these challenges, specifically in the online information landscape. CyberPeace actively publishes fact-checking reports and insights on Prebunking and Debunking, conducts expert sessions and takes various key steps aimed at empowering netizens to build cognitive defences to recognise the susceptible information, disregard misleading claims and prevent further spreads to ensure the true online information landscape.

References:

• https://www.scconline.com/blog/post/2024/09/20/bombay-high-court-it-rules-amendment-2023-fact-check-units-article14-article19-legal-news/#:~:text=Bombay%20High%20Court%3A%20A%20case,grounds%20that%20it%20violated%20constitutional
• https://indianexpress.com/article/cities/mumbai/bombay-hc-strikes-down-it-act-amendment-fact-check-unit-9579044/
• https://www.cyberpeace.org/resources/blogs/supreme-court-stay-on-centres-notification-of-pibs-fact-check-unit-under-it-amendment-rules-2023

‍

Introduction

YouTube is testing a new feature called ‘Notes,’ which allows users to add community-sourced context to videos. The feature allows users to clarify if a video is a parody or if it is misrepresenting information. The feature builds on existing features to provide helpful content alongside videos. Currently under testing, the feature will be available to a limited number of eligible contributors who will be invited to write notes on videos. These notes will appear publicly under a video if they are found to be broadly helpful. Viewers will be able to rate notes into three categories: ‘Helpful,’ ‘Somewhat helpful,’ or ‘Unhelpful’. Based on the ratings, YouTube will determine which notes are published. The feature will first be rolled out on mobile devices in the U.S. in English. The Google-owned platform will look at ways to improve the feature over time, including whether it makes sense to expand it to other markets. 

YouTube To Roll Out The New ‘Notes’ Feature

YouTube is testing an experimental feature that allows users to add notes to provide relevant, timely, and easy-to-understand context for videos. This initiative builds on previous products that display helpful information alongside videos, such as information panels and disclosure requirements when content is altered or synthetic. YouTube in its blog clarified that the pilot will be available on mobiles in the U.S. and in the English language, to start with. During this test phase, viewers, participants, and creators are invited to give feedback on the quality of the notes.

YouTube further stated in its blog that a limited number of eligible contributors will be invited via email or Creator Studio notifications to write notes so that they can test the feature and add value to the system before the organisation decides on next steps and whether or not to expand the feature. Eligibility criteria include having an active YouTube channel in good standing with Yotube’s Community Guidelines.

Viewers in the U.S. will start seeing notes on videos in the coming weeks and months. In this initial pilot, third-party evaluators will rate the helpfulness of notes, which will help train the platform’s systems. As the pilot moves forward, contributors themselves will rate notes as well.

Notes will appear publicly under a video if they are found to be broadly helpful. People will be asked whether they think a note is helpful, somewhat helpful, or unhelpful and the reasons for the same. For example, if a note is marked as ‘Helpful,’ the evaluator will have the opportunity to specify  if it is so because it cites high-quality sources or is written clearly and neutrally. A bridging-based algorithm will be used to consider these ratings and determine what notes are published. YouTube is excited to explore new ways to make context-setting even more relevant, dynamic, and unique to the videos we are watching, at scale, across the huge variety of content on YouTube.

CyberPeace Analysis: How Can Notes Help Counter Misinformation

The potential effectiveness of countering misinformation on YouTube using the proposed ‘Notes’ feature is significant. Enabling contributors to include notes on videos can offer relevant and accurate context to clarify any misleading or false information in the video. These notes can aid in enhancing viewers' comprehension of the content and detecting misinformation. The participation from users to rate the added notes as helpful, somewhat helpful, and unhelpful adds a heightened layer of transparency and public participation in identifying the accuracy of the content. 

As YouTube intends to gather feedback from its various stakeholders to improve the feature over time, one can look forward to improved policy and practical over time: the feedback mechanism will allow for continuous refinement of the feature, ensuring it effectively addresses misinformation. The platform employs algorithms to identify helpful notes that cater to a broad audience across different perspectives. This helps showcase accurate information and combat misinformation.

Furthermore, along with the Notes feature, YouTube should explore and implement prebunking and debunking strategies on the platform by promoting educational content and empowering users to discern between fact and any misleading information. 

Conclusion

The new feature, currently in the testing phase, aims to counter misinformation by providing context, enabling user feedback, leveraging algorithms, promoting transparency, and continuously improving information quality. Considering the diverse audience on the platform and high volumes of daily content consumption, it is important for both the platform operators and users to engage with factual, verifiable information. The fallout of misinformation on such a popular platform can be immense, and so, any mechanism or feature that can help counter the same must be developed to its full potential. Apart from this new Notes feature, YouTube has also implemented certain measures in the past to counter misinformation, such as providing authenticated sources to counter any election misinformation during the recent 2024 elections in India. These efforts are a welcome contribution to our shared responsibility as netizens to create a trustworthy, factual and truly-informational digital ecosystem.

References:

• https://blog.youtube/news-and-events/new-ways-to-offer-viewers-more-context/
• https://www.thehindu.com/sci-tech/technology/internet/youtube-tests-feature-that-will-let-users-add-context-to-videos/article68302933.ece

‍