Elsa Angel Rose
Life-like images, high-resolution Hollywood movie replica videos and animation videos, websites with articles about anything and everything, amidst the noise and online traffic of all the above, it has become an impossible task to identify what is synthetic (AI-generated) and non-synthetic information. An unasked-for but inevitable consequence of this AI boom is the unbridled spread of misinformation. Not saying that misinformation did not exist until AI entered human lives. But definitely saying that it has aggravated an already existing mammoth crisis of “Misinformation.” Especially during times of geopolitical wars, assembly elections, and more, being cognizant of this crisis is important. From many news reports of deep fakes and misinformation spreading, public figures, children, and women seem to be the easiest victims of the crime. In the last few months, Grok (an AI chatbot by Elon Musk) creating synthetic pornographic content; Seedance 2.0 (an AI video generator by ByteDance) producing a 15-second video of Brad Pitt and Tom Cruise fighting using just a two-line prompt; Sora (OpenAI’s video generator) introducing the Cameo feature, wherein users can create characters in videos resembling their real-life friends; and OpenAI’s ChatGPT effortlessly writing 1,000-word essays and news reports all make us ponder the importance of media and AI literacy. No more proof is needed to understand the capability of Artificial Intelligence, certainly way more capable than an average human. But in such times, in-depth media and AI literacy, along with awareness of the legalities, will help people avoid falling prey to misinformation or deepfake content and believing everything that AI has to offer. Initially, the dissemination of misinformation was the only crisis to be tackled during the advent of social media. But with the coming of AI, it has brought with it the problem of misinformation generation, adding to the already existing crisis of misinformation dissemination.
If you enjoyed reading this article, you can help support our journalism by adding OHeraldo as a preferred source.
AI KNOWS ITS GAME AND IS PLAYING WELL “Using modern AI tools, images, videos, and even voices can be generated to look extremely realistic. Earlier, fake content often had visible errors, but now AI can mimic real people’s writing styles, and other visuals also very convincingly,” shared Sachin Sutar from Velha. He is a responsible citizen who believes in verifying his sources of information and ensuring that he does not become part of the spread of misinformation. Regarding his daily experiences with AI, Sachin shared, “These days I often see videos online showing missiles being fired in different countries, politicians making shocking statements, or animals like lions attacking people riding bikes. They look very real at first. When I check the comments, many people say the content is AI-generated. Looking at the comments and discussions sometimes helps understand whether a post might be fake.” This instance is something to which many people agree; going to the comment sections where intense discussions are happening on whether the content is AI or real is an option opted for by many. “We’re at a genuinely strange moment. A year ago, AI-generated images and videos had tells, wrong hands, weird teeth, text that looked like someone sneezed on a keyboard. Today, mid range tools like Nano banana pro, FLUX, ChatGPT Images or Seedream produce images that most people would scroll past without a second thought,” said Raj Kunkolienkar, AI Consultant from Panaji. According to him, videos are also catching up fast. “Sora and similar models like Veo and Seedance now produce 30 to 5-minute clips that look like they were shot on location,” said Raj. When it is writing essays, reports or more, the bar is even higher. Raj explained, “A competent AI can write a 600-word news report in under a minute that reads clean. The average AI platform in 2026 is good enough to fool the average reader/viewer.” Fake news does not have to be complicated; it just needs to be fast-moving, which is one of the biggest strengths of AI.
HOW DO WE NAVIGATE THIS CRISIS?
Two solutions that come easily to mind are checking watermarks or getting the metadata of the content. But watermarks are not a relevant source of checking the authenticity of content; they can be removed, cropped and played with (there are many apps that offer this facility). But the latter option can be attempted. Metadata is information that is automatically attached to a piece of content at the time it is created. It provides additional details about how an image or video was produced. This can include information such as the type of camera used to capture a photo, the location, as well as the date and time a video was recorded, and the file name. Every photo or video contains metadata, regardless of whether it was created by a human or generated using AI. “For images and video, reverse image search is still your first line of defence. Google Images or TinEye will often show you where a photo first appeared and whether it’s been recycled from an older conflict. Tools like InVID or the Bellingcat verification toolkit go deeper, letting you check metadata and geolocation. When it comes to news, the discipline is simple but underused: go to the source. If a report claims a government said something, find the official statement,” said Raj. Things certainly get more interesting when the same AI can be leveraged by humans to verify the authenticity and source of different content. “Tools like Google’s SynthID embed invisible watermarks into AI-generated images. Platforms are using AI classifiers to flag deep fakes before they spread. News organisations like the BBC and AP now use detection tools in their verification workflows,” Raj Shared. He emphasised an arms race that we are in (metaphorically), just like how detection tools improve, generation tools also improve. Hence, the most durable option is not a technical fix, but media literacy, being mindful of what is happening around one, which are the credible sources that discuss the same, have confirmed the news and more.
BEING AWARE AND CAUTIOUS
High Court advocate Eeshan Usapkar from Mapusa explained, “In the present digital environment, individuals must exercise reasonable caution before accepting or sharing information. The authenticity of any content should be verified by checking the source, confirming whether it originates from a reputable media organisation or official authority, and cross-verifying the same information across credible platforms. In the case of images or videos, especially with the rise of AI-generated content, it is advisable to rely on recognised fact checking tools and official clarifications.” Regarding the support offered by the law, he shared, “Indian law provides remedies against the circulation of false or misleading information. The Information Technology Act, 2000 and the Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021, impose obligations on intermediaries to act against unlawful or misleading content. Further, depending on the nature and impact of the misinformation, provisions under the Bharatiya Nyaya Sanhita, 2023, may also be invoked where such content results in defamation, fraud, or public disorder.
