> It's critical to bring technologies like Veo to the world responsibly. Videos created by Veo are watermarked using SynthID, our cutting-edge tool for watermar

> It's critical to bring technologies like Veo to the world responsibly. Videos created by Veo are watermarked using SynthID, our cutting-edge tool for watermarking and identifying AI-generated content And we're supposed to believe that this is resilient against prompt injection? How do you prevent state actors from creating "proof" that their enemies engaged in acts of war, and they are only engaging in "self-defense"?