Google has launched a new music-generation feature in the Gemini app powered by DeepMind’s Lyria 3, a next-generation AI model designed to create original music from simple text prompts. Available to users aged 18 and over in English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese, the feature allows users to generate 30-second tracks with lyrics and accompanying cover art. Users can describe a mood, genre, or theme—such as a “comical R&B slow jam about a sock finding its match”—and Gemini will produce a unique composition. The tool also supports media-based prompts, enabling users to upload a photo or video and generate a track that matches its emotional tone or visual style. Lyria 3 represents an upgrade over earlier versions, producing more realistic and musically complex tracks with greater control over style, tempo, vocals, and instrumentation. Google emphasizes that the model is designed for original creative expression, not imitation. If a user mentions a specific artist in their prompt, Gemini interprets it as broad stylistic inspiration rather than a direct copy. The system includes safeguards to filter outputs against existing copyrighted content and prevent unauthorized replication of artists’ styles. To address concerns about transparency and misuse, all AI-generated music created with Lyria 3 is embedded with SynthID, Google’s imperceptible watermark that identifies content as AI-generated. This feature is now integrated into the Gemini app’s verification tools, allowing users to upload audio files and ask whether they were created using Google AI. Gemini will analyze the file for SynthID and use its own reasoning to determine the origin of the content. The feature is also expanding to YouTube, where creators can now use the Dream Track tool globally to generate AI music for videos. Previously limited to U.S. creators, this update makes the tool accessible worldwide. Google continues to collaborate with the music community, having launched initiatives like the Music AI Sandbox to explore ethical development and copyright compliance. Training for Lyria 3 has been conducted with careful attention to licensing agreements and intellectual property rights. Despite these efforts, AI-generated music remains controversial. While platforms like YouTube and Spotify are integrating AI music and signing deals with labels to monetize it, the industry faces ongoing legal challenges over the use of copyrighted training data. Some artists argue that AI models are trained on their work without consent. In response, companies like Deezer have introduced tools to label AI content and prevent fraudulent streaming. Google stresses that users must comply with its Terms of Service and Gen AI prohibited use policies, which ban violations of intellectual property and privacy rights. Users can report content they believe infringes on rights. Google AI Plus, Pro, and Ultra subscribers will receive higher generation limits. The rollout began on desktop and will expand to mobile devices in the coming days. Google’s goal is to make music creation accessible and fun, enabling users to add personalized soundtracks to everyday moments. The company sees this as part of its broader commitment to responsible AI development, balancing innovation with accountability. As AI continues to reshape creative industries, tools like SynthID and transparent generation practices are becoming essential for maintaining trust and integrity in digital content.