At the Google I/O 2025 event on May 20, Google announced the release of Veo 3, a new AI video generation model that creates 8-second videos. Within hours of release, AI artists and filmmakers were showing off shockingly realistic videos. You may not even have seen some of these videos on your social media feed and realize that they are artificially generated.
To dull: I've never seen anything like Veo 3. It's impressive. It's scary. And it's just to get Better.
Misinformation experts have been warning us for years that we will eventually reach the point where it is impossible for the average person to convey the difference between AI videos and the real thing. In VEO 3, we officially emerged from the creepy valley and came out in a new era where AI videos are the facts of life.
There are several other AI video makers, especially Sora from Openai, but the clips created by Veo 3 instantly stand out in the timeline. VEO 3 has brought some innovations that separate it from other video generation tools. Importantly, in addition to the video, VEO 3 also generates audio and dialogs. Not only does it provide photorealism, it also provides soundscapes and conversations that are fully realized for video. It also allows users to maintain a consistent character across a variety of video clips, allowing users to fine-tune camera angles, framing, and movements in a whole new way. On social media, many users are umbing up with results.
This tweet is currently unavailable. It may have been loaded or deleted.
VEO 3 is currently available on Google's paid AI plan. Users have access to the tools of Gemini, Google's AI chatbot, and Flow has access to Google with “by creative, AI filmmaking tools built for creatives.”
Already, AI filmmakers have used VEO 3 to create short films, but it's only a matter of time before they see full-length films with VEO 3.
Meet filmmakers who make short films in VEO 3
On X, YouTube, Instagram and Reddit, users share some of the most impressive VEO 3 videos. If you're not in your security guard, you may never again think about whether the video is real or not, if you simply casually scroll through your feed.
The short film “Inflenders” is one of the most widely shared short films made with VEO 3. “Inflenders” was created by Yonatan Dor, founder of AI Visual Studio The Dor Brothers. In the film, an inexplicable change occurs in the background, resulting in a series of influencers responding. This video has hundreds of thousands of views across a variety of platforms.
“Yes, we used Google Veo 3 exclusively for this video, but we had to do more sound design, clever editing and some upscaling to make this work really come true,” Dor said in an email to Mashable. “The full piece took about two days to complete,” DOR said, “VEO 3 is a massive step forward and the most sophisticated tool available to the public today. We are particularly impressed by its dialogue and rapid adherence capabilities.”
Masculine light speed
Artists like Alex Patrascu and Impekable show off their Veo 3 abilities, and similar videos featuring street videos have gone viral. And at the beginning of this week, Wall Street Journal The reporter used VEO 3 to create an entire short film starring his own virtual version. This all took just 10 days.
In “Influencer” and these other videos, some clips and characters are more realistic than others. Many still have the shiny aesthetic and jerky camera movement that is the signature of AI video.
Just a few years ago, AI works with too many fingers and other obvious anatomical abnormalities were common. As technology continues to progress at this rate, there is no obvious difference between actual video and AI video.
Tools for creativity and misinformation
When promoting VEO 3, Google wants to highlight its partnerships with artists and filmmakers like Darren Aronofsky. And it's clear that VEO 3 can significantly reduce the cost of creating animations and special effects. But for content farms and bad actors who produce fake news and manipulative rage baits, VEO 3 is equally powerful.
Asked Google about the possibility that VEO 3 could be used for misinformation, the company said protections such as digital watermarks are built into the VEO 3 video clips.
“It's important that people have access to source tools for videos and other content that they can watch online,” a Google DeepMind representative told Mashable in an email. “The SynthID watermark is embedded in all the content generated by Google's AI tools, and the SynthID detector was deployed to testers early last week. We are planning to expand access more widely soon, and we will add a visible watermark to our VEO videos as an additional step to help people.”
Google also has safety guidelines for the AI used, which the company says “helps AI to responsibly create and identify content generated by it.”
Screenshots of video generated in AI created by Google using Google 3.
Credit: Google
But will the average person stop asking if the timeline and FYP images and videos are real? They don't, as kangaroos prove the emotional support of the virus.
There is no doubt that AI videos will become even more common on social media and video apps. This includes many AI slops, but also videos with more evil purposes. Despite the protective guard built into AI video generation tools, skilled AI artists can create Deepfake videos featuring celebrities and public figures. TV News anchors that are exposed to cameras have been a recurring theme in VEO 3 videos so far, and have had a worrying impact on the online information ecosystem.
If you haven't yet asked “Is this true?”, when you come across a video clip online, it's time for you to begin.
Or as if the chorus of voices were X saying, “We're so cooked.”
Follow Timothy Beck Werth (@beck_werth)and Mashable (@mashable) Latest AI News X and analysis.
Disclosure: Mashable's parent company Ziff Davis filed a lawsuit against Openai in April, claiming it infringed Ziff Davis's copyright in training and operating AI systems.
