I Cloned Myself With Gemini’s AI Avatar Tool. The Result Was Unnervingly Me (www.wired.com)

Google has launched a new feature in its Gemini app that allows users to create highly realistic AI avatars, offering a glimpse of what digital self-representation can look like. Using a process that requires minimal setup—just a smartphone camera and a few prompts—users can generate lifelike videos featuring their avatars in various scenarios. The reviewer created clips of their AI clone singing to a dinosaur and surfing under the Golden Gate Bridge, showcasing the balance of impressive realism and unexpected absurdity in the generated content. This innovation is significant for the AI/ML community as it pushes the boundaries of personalization and virtual interaction, reminiscent of OpenAI’s Sora app but with improvements. While the avatar feature has potential for creative applications, it also raises concerns over misuse, particularly with regards to nonconsensual deepfakes. Google emphasizes its commitment to safety, aiming to prevent harm without stifling creative possibilities. With applications like these, the line between digital and real-world identities continues to blur, prompting a need for ethical considerations in AI development.