Last month, Google launched Veo 3, its most advanced AI video generator to date. Producing clips so convincing they’re often indistinguishable from real footage, Veo 3 is being hailed as a breakthrough in synthetic media. But while some are embracing the tool’s creative potential, Veo’s realism signals a new stage in the eroding boundaries between truth and simulation.
Unlike earlier tools riddled with glitches and uncanny visuals, Veo is capable of producing hyper-realistic video clips with cinematic visuals, synchronized dialogue and sound effects. From eerily lifelike monologues to full-blown short films, users are flooding social media with Veo-generated clips that are often indistinguishable from real footage.
The tool’s near-flawless fidelity raises concerns about the increased plausibility of deepfakes. In tests by journalists and watchdogs, Veo 3 has also shown it can generate realistic, misleading content. While Google has added watermarking and safety filters, experts warn that the tool’s outputs are already circulating online in ways that could spark real-world harm.
For more like this, subscribe to our Digital Insights newsletter.