By Bindiya Bhatt
The feature image above looks like a picture of a real monkey playing tabla, right? But wait! What you are seeing is actually a screengrab from an AI-generated video made using the newly launched Google tool Veo 3. From the colourful ethnic jacket worn by the monkey to the smiling ladies in the backdrop, everything about this picture is artificial. Indian tech experts are terrified as this picture looks so real that it can cause damage, indicating how things can be so realistic as to create confusion among the people.
Experts are worried that AI tools like Veo 3 could have more dangerous effects than previously thought. They may speed up the spread of misinformation and blur the line between what’s real and what’s fake. Previously, one could spot an AI video in several ways — the clip could contain discrepancies like a person with three hands, legs, or arms. But with technological advancements in such models, these signs are now fading.
To understand the risks, challenges, and what India must prepare for, we spoke to AI expert Devika Chhibber Mehta, who started by explaining what exactly Google’s Veo 3 is, and why is it different from earlier video tools?
“Veo 3 is an AI model that uses straightforward text or image prompts to create lifelike video clips with built-in sound. Its greater simulation, synchronised dialogue, background music, and sound effects within the video are what distinguish it from previous tools,” she said.
When asked how serious the risk is of such videos being used to spread false news or fear in India, she pointed to the country’s vast population, high internet penetration, and low AI literacy as key factors driving the spread of such content.
“If Google has created Veo 3, it has also implemented watermarking policies and has launched an AI-generated detection tool, Synth ID. But with India’s vast population, high internet penetration and lower AI literacy levels, it is easy to spread such malicious content, destabilising public order,” she said.
Asked about the strength of current laws in India, especially during sensitive times like elections or protests, she said more needs to be done to combat deepfakes.
“Across the globe, countries are trying to counter deepfakes and it is a bigger menace. India is also trying to regulate; we have existing provisions under the IT Act and IPC for offences like defamation or impersonation, but more is required.”
When asked whether it is possible for a regular person to tell the difference between a real video and one made by AI and what people should look out for she said, “On social media, AI-generated content is frequently produced using free AI tools, which might not look very natural. You can tell by looking for irregularities in the image, such as out-of-sync lip movements, blurry text and head movements.”
She also suggested a tool that can identify fake videos generated using AI before they go viral. “One can try Hive Moderation, as it gives you some free credits to get started.”
Devika concluded by outlining steps the government, tech companies, and even we as citizens should take to prevent the misuse of AI-generated videos.
“The tech companies must create strong detection tools and enforce more stringent platform policies while simultaneously encouraging AI literacy; the government must ensure that these policies are strictly followed. As citizens, it is our responsibility to actively report suspicious deepfakes, critically assess online content, and confirm information from reliable sources. Always verify before you share any information within your circles,” she said.
Google Drops ‘Bombshell’
Google recently dropped a ‘bombshell’ by launching an advanced AI-powered video generation tool, Veo 3, which is now available to users in India. A report published on TIME recently analysed the tool, and the outcome was terrifying.
According to the publication, the AI tool can “generate realistic clips” containing misleading or explosive information about news events. TIME reported that the videos created using Veo 3 included shocking scenes such as “a Pakistani crowd setting fire to a Hindu temple; Chinese researchers handling a bat in a wet lab; an election worker shredding ballots; and Palestinians gratefully accepting US aid in Gaza.”
Real or Fake? Confusion Already Reigns
It is becoming increasingly difficult to tell whether a video is real or fake, with incidents of real clips being mistaken for AI-generated ones going viral on social media.
Recently, a post on X went viral after claiming that a Daily Wire journalist had shared an AI-generated video of aid being distributed in Gaza. However, a BBC journalist later confirmed the footage was authentic.
In contrast, a video of an “emotional support kangaroo” trying to board an airplane also went viral, with many users convinced it was real. However, it turned out to be AI-generated.
The Fake News Experiment: How Easily Can We Be Fooled?
After the launch of Veo 3, internet researcher Henk van Ess conducted a fake news experiment using the tool, fabricating a political scandal from scratch. He stitched together short video clips and edited them into a fake newsreel claiming that a yacht manufacturing facility would replace a small-town school.
I decided to use my lunch time to show you how easy it is to make a fake news story in 30 minutes with Veo3 (I didn't try to perfect it). First: the footage. A mayor comes with a crazy idea and people hate it: (1/10) #verification #ai pic.twitter.com/kHF0lqaR8i
— 𝚑𝚎𝚗𝚔 𝚟𝚊𝚗 𝚎𝚜𝚜 (@henkvaness) May 27, 2025
“If I can create one convincing fake story in 28 minutes, imagine what dedicated bad actors can produce,” he wrote on Substack. “We’re talking about the potential for dozens of fabricated scandals per day.”
How scary can this technology be!
Google Responds To Concerns
Despite the risks associated with the tool, Google says it has policies in place to protect users and is committed to responsible AI development.
“Veo 3 has proved hugely popular since its launch. We’re committed to developing AI responsibly and we have clear policies to protect users from harm and governing the use of our AI tools,” a Google spokesperson said in a statement.
“We’ve also added visible watermarks to all videos, except for videos from Ultra members in Flow. This is a first step as we work to make our SynthID Detector available to more people in parallel,” Josh Woodward, Vice President with Google Labs and Google Gemini, informed on X.
Lots more coming, and we're tracking your top requests:
— Josh Woodward (@joshwoodward) May 30, 2025
* More countries (India, Indonesia, EU, etc.)
* Image-to-video on @GeminiApp
* More reliable audio generation
* Faster generations
* Support for Workspace accounts
Let us know if there are other things you want to see!
Veo’s website states that it blocks “harmful requests and results.”
Google has also released a technical paper alongside Veo 3, which says Veo 3 “is generally prone to small hallucinations that mark videos as clearly fake. Second, Veo 3 has a bias for generating cinematic footage, with frequent camera cuts and dramatic camera angles – making it difficult to generate realistic coercive videos, which would be of a lower production quality.”
ALSO READ: Nostalgia Mode On! Mario Bros To Prince of Persia – My Favourite Retro Games From The ’80s
