Strings of Reality

MAKING THE PUPPETS:

Deepfakes are media that are digitally manipulated to replace a person's likeness convincingly with that of another. Deepfakes are the manipulation of facial appearance through deep generative methods.[1] Synthetic media (also known as AI-generated media, media produced by generative AI, personalized media, personalized content and colloquially as deepfakes) is an umbrella term consisting of AI production, manipulation, and modification of data and media through automated means, primarily through the use of artificial intelligence algorithms, such as to mislead people or change an original meaning.[2]

Deepfake creation ML (machine-learning) methods have foundations in deep learning concepts and consist of neural network architecture training, more importantly, the generative ones, like generative adversarial networks (GANs).

PUPPET THEATER:

Deepfake caught the world off-guard in 2017 when adult videos of celebrities were posted on Reddit by an individual user.[3] From that moment, the world of fake news and synthetic media has just skyrocketed. An abomination earlier used to target celebrities and politicians gradually found another use to target innocent people for fraud and scams. Virat Kohli was the most recent target as his face was used to promote a betting app [4], Mukesh Ambani promoted ‘Free Stock Forum’ in another video [5], and who can forget Rashmika Mandanna’s deepfake video, which caused a stir?[6]

The list is long and updates daily with the boost of AI technology and its availability to the masses. At the advent of the upcoming Lok Sabha elections, Google has decided to join hands with Indian fact-checkers to fight deepfakes[7]. Techies excelling in deepfake technology are in demand for various political campaigns of different political parties for the 2024 elections.[8]

Scams and frauds using deepfake technology are the latest bane, with a company losing Rs. 207 crores due to a sophisticated deepfake scam [9]. This has also been integrated into scamming individuals, with an older man from Kerala failing to distinguish the voice of a former colleague from an AI-enabled voice and losing 40 thousand rupees [10].

Deepfake tech is being used to blackmail innocent people by creating deepfakes of them doing illegal and immoral actions and later blackmailing them. An example can be the case in Ghaziabad, where a retired IPS officer’s face and voice were replicated using deepfake tech to blackmail and extort money from an innocent senior citizen [11].

POSSIBLE PERFORMANCES:

Deepfake technology is a storm that can be used for and by the good but will destroy everything if let into the hands of the bad. It contains many ways to dismantle the world and is still evolving and improving. The quality of deepfakes keeps improving; thus, distinguishing them from the original gets way tougher.

Some implications may include –

● Disinformation Campaigns: 

Deepfakes are fake impressions of a person, and this further can be used to spread disinformation about someone on a mass level. It can be done in many ways; for example, someone can be shown saying something offensive or doing something immoral. Information is essential as even a tiny amount of it can damage a person's reputation.

● Political Misuse:

Deepfakes of politicians and damaging their reputations aren’t new news, but this, in turn, undermines public trust in the government, increases social divisions and eventually manipulates public opinion on a large scale. Malicious use of deepfake is also a national security risk as central officials can be the targets.

● Journalistic Ethics:

Objectivity of news diminishes with the rise of deepfakes as anyone can create media that supports their story and fuels their revenue. The line between real and fake is skinny, which poses a severe challenge to ethical journalists. Public trust in journalism and media also is harmed due to deepfakes and fake news.

● Personal Privacy and Repute:

Anyone can be targeted. It’s as simple as that. Someone may wake up to find themselves viral on the Internet in an unwanted manner. Someone may wake up to blackmailing messages and scam calls. This is the truth of today.

These are some possible ‘performances’ of the puppet play that is deepfake.

NO-SHOW:

The new age of AI has brought new waves of crime and new ways to fight them. The world is realizing the dangers and taking relevant actions to curb them and face these challenges.

Google is formulating a policy to increase awareness and educate YouTube creators about deepfake and how to tackle it. The policy will focus on regulating the use of AI-generated content on platforms like YouTube. According to the company, the policy focuses on creator disclosures when it alters reality and mandates labelling of electoral advertisements crafted by GenAI, among others, besides enabling creators to watermark their content using Google’s tools [12]. 

MCA and Meta collaborate to fight AI-enabled fake information/ misinformation using helpline on WhatsApp. The Misinformation Combat Alliance will set up a Deepfakes Analysis Unit to identify and curb the spread of AI-enabled disinformation among internet users in India [13].

The synthesis and manipulation of Twitter's media policy is about deepfake technology-enabled content and media shared on the platform [14].

The Indian Government is set to notify rules that will mandate social media platforms to tackle the menace of deepfakes after repeated advisories failed to get the desired result, Union Minister of State for Information Technology Rajeev Chandrasekhar said. “We issued an advisory, and we also said that if we are not satisfied with the compliance, we will notify newly amended rules that are much more specific to the issue of misinformation and deepfake in particular. You can take it from me today that we will issue newly amended IT rules over the next seven to eight days," the minister states. “If it is not enough after the government suggests through an advisory what amendments have to be done to the Terms of Use, we will then embed that explicitly in the rules, which we’re now going to do," he said[15].

These steps show us hope that the harmful use of deepfake technology will be curbed and that we will all head towards a better future.

CONCLUSION-

This puppet show of deepfake is very harmful and a no-go, but we cannot just reject it. We must become aware and media-literate and empower ourselves to fight against this threat.

CITATIONS –

[1] - Wikipedia Contributors. (2019, May 17). Deepfake. Wikipedia; Wikimedia Foundation. https://en.wikipedia.org/wiki/Deepfake

[2] - Synthetic media. (2023, January 31). Wikipedia. https://en.wikipedia.org/wiki/Synthetic_media

[3] - Roettgers, J., & Roettgers, J. (2018, February 21). Porn Producers Offer to Help Hollywood Take Down Deepfake Videos. Variety. https://variety.com/2018/digital/news/deepfakes-porn-adult-industry-1202705749/

[4] - HT Sports Desk. (2024, February 20). Virat Kohli’s “deepfake video” goes viral: AI-generated clip shows ex-India skipper promoting betting app. Hindustan Times. https://www.hindustantimes.com/cricket/virat-kohli-deepfake-video-goes-viral-ai-generated-clip-shows-ex-india-skipper-promoting-betting-app-ind-vs-eng-101708431997959.html

[5] - Mehrotra, K. (2024, March 4). Fact-Check: Video of Mukesh Ambani Promoting “Free Stock Forum” Is a Deepfake! TheQuint. https://www.thequint.com/news/webqoof/deepfake-video-of-reliance-mukesh-ambani-promoting-stocks-fact-check

[6] - Ojha, A. (2023, November 10). Rashmika Mandanna deepfake video: Delhi Police registers case. India Today. https://www.indiatoday.in/india/story/rashmika-mandanna-deepfake-video-delhi-police-case-registered-2461547-2023-11-10

[7] - Gupta, A. (2024, March). Google partners with news publishers to curb deepfakes ahead of elections. @Bsindia; Business Standard. https://www.business-standard.com/elections/lok-sabha-election/google-partners-with-news-publishers-to-curb-deepfakes-ahead-of-elections-124030100801_1.html

[8] - Dhawan, H. (2024, February 29). Netas will fake it till they make it. The Times of India. https://timesofindia.indiatimes.com/india/the-rise-of-deepfakes-in-indian-politics-ai-powered-manipulation-in-elections/articleshow/108087728.cms

[9] - Garg, A. (2024, February 5). Deepfake scam: Company loses around Rs 207 crore after employee connected to a video call. India Today. https://www.indiatoday.in/technology/news/story/deepfake-scam-company-loses-around-rs-207-crore-after-employee-connected-to-a-video-call-2497996-2024-02-05

[10] - Varma, V. (2023, July 18). Kerala man loses ₹40k to AI-enabled deep-fake fraud. Hindustan Times. https://www.hindustantimes.com/india-news/deepfake-scammers-trick-indian-man-into-transferring-money-police-investigating-multi-million-rupee-scam-101689622291654.html

[11] - Kumar, A. (2023, November 30). Sextortionists target senior citizens with retd IPS officer’s deepfake. The Times of India. https://timesofindia.indiatimes.com/city/ghaziabad/retired-ips-officers-deepfake-used-to-blackmail-senior-citizen/articleshow/105603283.cms

[12] - Das, G. A. (2023, November 29). Google to tackle deepfakes the right way: VP Saikat Mitra. Mint. https://www.livemint.com/companies/news/google-to-tackle-deepfakes-the-right-way-says-v-p-mitra-11701278222484.html

[13] - PTI. (2024, February 19). Crackdown on deepfakes: MCA, Meta partner to curb AI-generated misinformation with WhatsApp helpline. The Hindu. https://www.thehindu.com/sci-tech/technology/crackdown-on-deepfakes-mca-meta-partner-to-curb-ai-generated-misinformation-with-whatsapp-helpline/article67863420.ece#:~:text=The%20collaboration%20with%20MCA%20to

[14] - Twitter. (2023, April). Synthetic and manipulated media policy. Help.twitter.com. https://help.twitter.com/en/rules-and-policies/manipulated-media

[15] - Aulakh, G. (2024, January 16). Govt set to notify new IT rules on deepfakes for social media: Chandrasekhar. Mint. https://www.livemint.com/news/india/why-fogs-disrupt-flights-in-india-every-year-and-how-to-fix-it-11705426106615.html