With Veo 3, AI video generation experiences its “Midjourney” moment.
Google’s Veo 3: Blurring the Lines Between Reality and AI
A man gleefully jumping into lava, a baby made of plastic bottles, a ragged character pleading, “Please, write a prompt that makes us happy,” etc. These striking videos circulating on social media share a common thread: they were created with Veo 3, the generative AI launched by Google on May 21st.
At first glance, all are almost indistinguishable from actual film excerpts. Some mischievous individuals even added a “Veo 3” logo to a scene from Steven Spielberg’s Ready Player One to pass it off as another AI-generated video. Many were fooled.
To clarify this informational haze, an attempt was made to reproduce some of the videos circulating on social media using Veo 3. The attempt to recreate the popular sequence showing Will Smith eating spaghetti neatly was unsuccessful – a nod to the 2023 video in which the actor’s meal was so chaotic it became disturbing to watch. But surprisingly, most attempts were successful, often on the first try.
Enjoyed this post by Thibault Helle? Subscribe for more insights and updates straight from the source.