Google introduced Gemini on December 6, 2023, as its first natively multimodal AI model, designed from the ground up to seamlessly understand and operate across text, images, audio, and video, unlike previous models that stitched together separate components.