PM Modi sounds alarm on deepfakes: How can you spot them?

Prime Minister Narendra Modi has sounded the alarm on deepfakes. Modi, speaking at the BJP’s ‘Diwali Milan’ programme at the party’s headquarters in Delhi on Friday, said he recently saw a deepfake of himself singing at a garba festival. Modi said he has not done so since he attended school. “A new crisis is emerging due to deepfakes produced through artificial intelligence. There is a very big section of society which does not have a parallel verification system,” the prime minister said. Modi’s remarks come in the backdrop of a series of deepfakes of Rashmika Mandanna, Katrina Kaif and Kajol going viral on social media. The Centre had last week issued an advisory to major social media companies to identify misinformation, deepfakes and other content that violates rules and remove those within 36 hours after being reported. But what is a deepfake? And how do you spot it? Let’s take a closer look: What are they? Deepfakes are manipulated media in which a person’s likeness – either in an image or a video – is replaced with someone else. As per The Guardian, they are this century’s version of photoshopping. Deepfakes are created via Artificial Intelligence using deep learning algorithms, as per Indian Express.

The word itself is a portmanteau of deep learning and fake.

They are made by feeding a computer an algorithm, or set of instructions, lots of images and audio of a certain person. The computer program learns how to mimic the person’s facial expressions, mannerisms, voice and inflections. If you have enough video and audio of someone, you can combine a fake video of the person with a fake audio and get them to say anything you want. Most deepfakes cannot be created on standard PCs or laptops. As per The Guardian, they are made on high-end PCs with powerful graphics cards or using cloud computing. Deepfakes came to in 2017 when a Reddit user put the face of celebrities – sounds familiar? – such as Scarlett Johanson, Gal Godot and Taylor Swift on porn actresses. While criminal gangs are using deepfakes to con people out of their hard-earned money, deepfakes have also been used to target and harass women via revenge porn. In the United States, deepfakes of Hillary Clinton and Joe Biden endorsing Republican Ron Desantis and abusing a trans person respectively – have gone viral on social media blurring fact and fiction in the already polarised world of US politics. How do you spot it? While major social media platforms like Facebook, Twitter, and YouTube have made efforts to prohibit and remove deepfakes, their effectiveness at policing such content varies. Experts say deepfakes are not easy to spot. However, MIT spells out a few tell-tale signs including:

1. Looking at the person’s face particularly the cheeks, forehead, eyes and eyebrows.
2. Have a look at the glasses in the clip.
3. Keep your eyes peeled for unrealistic looking facial hair
4. Is the person in the video blinking too little or too much?
5. Check out lip movements. Are they in sync? Do they look normal?

As per Indian Express, you can also use AI detection tools that are available online. More importantly, it is important to keep yourself up-to-date with current affairs and not share a clip unless you are absolutely sure it is legitimate. A scam that is gaining in popularity is getting a video or audio call from a friend or loved one in dire need of money. A former Coal India employee lost Rs 40,000 after getting such a video call from a friend. As per ZDNet, experts say one way to check if such a caller is genuine is by simply asking to turn their face to one side.

This is because deepfake AI models are not yet proficient at recreating side-on views.

The website quoted the FBI as saying that sounds that don’t concur with what is happening in the video – a cough or a sneeze for example are also a giveaway. The website quoted Martin Anderson as writing for Metaphsyics.ai, ““That paucity of available data makes it difficult to obtain a range of profile images on non-celebrities that’s diverse and extensive enough to train a deepfake model to reproduce profile views convincingly.” “This weakness in deepfakes offers a potential way of uncovering ‘simulated’ correspondents in live video calls, recently classified as an emergent risk by the FBI: if you suspect that the person you’re talking to might be a ‘deepfake clone’, you could ask them to turn sideways for more than a second or two, and see if you’re still convinced by their appearance.” ‘New crisis emerging’ Modi said in a diverse society like ours, deepfakes can cause a big crisis and even stoke the fire of disaffection in the society as people generally trust anything associated with the media in the same way anyone clad in “gedua” (saffron) tends to get respect from others. “A new crisis is emerging due to deepfakes produced through artificial intelligence. There is a very big section of society which does not have a parallel verification system,” the prime minister told journalists at the programme where he aired views about a few things and sought their support to raise awareness. The prime minister noted that earlier a movie with some controversial comments would come and go, but now it goes on to become a big issue. Even screening of such films becomes difficult on the ground that they have disrespected some section of society irrespective of the huge amount of money spent on making them, he noted. [caption id=“attachment_13397882” align=“alignnone” width=“640”] Taking cognisance of how dangerous deepfakes can be, Prime Minister Narendra Modi has expressed grave concerns over the misuse of AI and has cautioned people to take precautions against falling for AI-generated fake videos[/caption] Modi suggested that as products like cigarettes come with health warnings, deepfakes too should carry disclosures. ‘Call upon platforms to take decisive action’ Minister of Electronics and IT Rajeev Chandrasekhar last week said that deepfakes are a major violation and harm women in particular. “Safety and trust of our Digital Nagriks is our unwavering commitment and top priority for the Narendra Modi Government. “Given the significant challenges posed by misinformation and deepfakes, the Ministry of Electronics and Information Technology (MEITY) has issued a second advisory within the last six months, calling upon online platforms to take decisive actions against the spread of deepfakes,” Chandrasekhar said. He said that the government takes the responsibility of safety and trust of all citizens very seriously, and more so about our children and women who are targeted by such content. “It is a legal obligation for online platforms to prevent the spread of misinformation by any user under the Information Technology (IT) rules, 2021. [caption id=“attachment_13396362” align=“alignnone” width=“640”] After Rashmika Mandanna and Katrina Kaif, a deepfake video based on Kajol, changing into an outfit has gone viral on social media. Given how cheap and easy it has become to work with AI and create your own bots, such deepfakes will only become more common[/caption] “They are further mandated to remove such content within 36 hours upon receiving a report from either a user or government authority. Failure to comply with this requirement invokes Rule 7, which empowers aggrieved individuals to take platforms to court under the provisions of the Indian Penal Code (IPC),” the minister said. “The Centre today issued an advisory to the significant social media intermediaries to ensure that due diligence is exercised and reasonable efforts are made to identify misinformation and deepfakes, and in particular, information that violates the provisions of rules and regulations and/or user agreements,” the Ministry of Electronics and IT (Meity) said in a statement on 7 November. Meity in the advisory has directed that social media companies expeditiously take action against the content that is in violation of the rules and regulations well within the timeframes stipulated under the IT Rules 2021, and users are caused not to host such information, content, deepfakes. “Remove any such content when reported within 36 hours of such reporting and ensure expeditious action, well within the timeframes stipulated under the IT Rules 2021, and disable access to the content or information,” the statement said. In case of impersonation in an electronic form, including artificially morphed images of an individual, social media companies have been advised to take action within 24 hours from the receipt of a complaint in relation to any content. “The intermediaries were reminded that any failure to act as per the relevant provisions of the IT Act and Rules would attract Rule 7 of the IT Rules, 2021 and could render the organisation liable to losing the protection available under Section 79(1) of the Information Technology Act, 2000,” the statement said. With inputs from agencies