When you buy through affiliate links in our content, we may earn a commission at no extra cost to you. Learn 
Google unveiled its state-of-the-art text-to-video AI model, Veo, at Google I/O in May 2024. The model can generate 1080p video clips of over 60 seconds in length in over 100 cinematic styles, including photorealism, surrealism, and animation.
Veo is available to try via the experimental tool VideoFX and belongs to an emerging category of generative AI text-to-video models that aim to offer video production capabilities to non-technical users and artists.
The platform’s initial examples have been impressive so far. They provide serious competition against OpenAI’s video generation model Sora, which was announced at the start of this year to significant hype.
In this article, we’ll examine some of the best videos made with Google Veo that we’ve seen so far.
Key Takeaways
6 Awesome AI-Generated Videos Made with Google Veo
1. Veo
To help prompt the release of Veo, Google DeepMind released a short video depicting a neon-lit city. The video begins with the camera panning over the city’s skyline before diving down to street level and following a car traversing the streets.
The detail in the video is incredible, from the motion of the car travelling through the streets down to the wet look of the road and the mist catching the edge of the neon lights.
On first viewing, the only natural part we noticed was the awkward transition from night to day in the tunnel. That being said, the daylight depiction of Hong Kong appeared highly realistic, even with the inconsistency between the car’s motion and the movement of road markings.
Introducing Veo: our most capable generative video model. 🎥
It can create high-quality, 1080p clips that can go beyond 60 seconds.
From photorealism to surrealism and animation, it can tackle a range of cinematic styles. 🧵 #GoogleIO pic.twitter.com/6zEuYRAHpH
— Google DeepMind (@GoogleDeepMind) May 14, 2024
Prompt used to create the video:
“A fast-tracking shot through a bustling dystopian sprawl with bright neon signs, flying cars and mist, night, lens flare, volumetric lighting.
A fast-tracking shot through a futuristic dystopian sprawl with bright neon signs, starships in the sky, night, volumetric lighting.
A neon hologram of a car driving at top speed, speed of light, cinematic, incredible details, volumetric lighting.
The cars leave the tunnel, back into the real world city Hong Kong.”
2. Spotted Jellyfish
Our next example is a video of a jellyfish swimming in the ocean. While this is just a short clip, everything from the motion of the jellyfish in the water to the ripples of the water appears authentic and natural.
At a glance, it would be difficult to identify that this video wasn’t shot with a traditional camera, as the spots and texture of the jellyfish look so lifelike. It was also created with a simple text prompt.
✍️ Prompt: “Many spotted jellyfish pulsating under water. Their bodies are transparent and glowing in deep ocean.” pic.twitter.com/y9SmNd8NK0
— Google DeepMind (@GoogleDeepMind) May 14, 2024
Prompt used to create the video:
“Many spotted jellyfish pulsating under water. Their bodies are transparent and glowing in the deep ocean.”
3. Golden Retriever
Another example released by DeepMind features a golden retrieval traversing a mountain trail. This video stands out among the examples due to how real the dog looks, particularly with regard to how it walks and wags its tail (although the motion of the fur looks unnatural around the back legs).
The video background also appears almost indistinguishable from real footage, with the trail appearing to slope down the mountain as it would in a legitimate physical location. Overall, nothing appears out of place or noticeably deepfaked, meaning the image could easily pose as a real video.
✍️ Prompt: “A golden retriever walks on a winding mountain trail, its tail wagging excitedly as it explores the sights and scents of the wilderness.” pic.twitter.com/CeJ5VKfAaM
— Google DeepMind (@GoogleDeepMind) May 14, 2024
Prompt used to create the video:
“A golden retriever walks on a winding mountain trail, it’s tail wagging excitedly as it explores the sights and scents of the wilderness.”
4. The Woman in the Cafe
Deepmind’s film noir-style promotional video, which features a woman sitting in a cafe, does a great job of depicting human beings.
The woman depicted in the video appears to be extremely lifelike in terms of her facial expression and demeanor. That being said, it’s easy to notice some unnatural motion in the fingers as they grip the book she’s holding.
While the video is quite short, it could easily be used as a B-roll for a creative project or film and illustrates the possibilities that arise from using this technology.
✍️ Prompt: “A woman sitting alone in a dimly lit cafe, a half-finished novel open in front of her. Film noir aesthetic, mysterious atmosphere. Black and white.” pic.twitter.com/vFVXr4Cvxi
— Google DeepMind (@GoogleDeepMind) May 14, 2024
Prompt used to create the video:
“A woman sitting alone in a dimly lit cage, a half-finished novel open in front of her. Film noir aesthetic, mysterious atmosphere. Black and white.”
5. Water Lily Timelapse
One of the more unique examples among DeepMind’s promotional videos is a short clip of a water lily slowly opening. In the video, the water lily gradually opens as if it were part of a time-lapse.
The motion of the lily opening is extremely detailed and convincing, even though the background appears noticeably lifeless and dark.
This example highlights how tools like Google Veo can be used not just for typical video content but could also be used to create time lapses and provide a visual breakdown of certain processes.
✍️ Prompt: “Timelapse of a water lily opening, dark background.” pic.twitter.com/t5uLQ89E1Y
— Google DeepMind (@GoogleDeepMind) May 14, 2024
Prompt used to create the video:
“Time-lapse of a water lily opening, dark background.”
6. Kebabs Grilling
Our final example is a short clip that shows AI-generated chicken and green pepper kebabs on a grill. One of the standout elements of this clip is just how natural the movement of the fire and the smoke looks.
However, our only criticism is that the angle and focus of the camera appear unnatural i,e., it doesn’t look like it was recorded with a camera.
✍️ Prompt: “Extreme close-up of chicken and green pepper kebabs grilling on a barbeque with flames. Shallow focus and light smoke. vivid colours.” pic.twitter.com/LDHC8XGyJA
— Google DeepMind (@GoogleDeepMind) May 14, 2024
Prompt used to create the video:
“Extreme close-up of chicken and green pepper kebabs grilling on a barbecue with flames. Shallow focus and light smoke. vivid colours.”
The Bottom Line
2024 is the year of the AI-generated video, with all the major players stepping their way into the game — something that has creatives both intrigued and worried.
Veo is another impressive entry into the text-to-video market, and DeepMind’s initial promotional videos illustrate that it has the core capabilities necessary to create high-quality video content.
We are still only months into this new industry — and we wait to see what comes next.
