<\/p>\n
Universal Music Group and Sony Music Entertainment (Sony) want to update a lawsuit against the AI music-generation platform Suno.<\/p>\n
As AFROTECH\u2122<\/a> previously reported, major<\/span>\u00a0<\/span>record labels<\/a>\u00a0<\/span>UMG and Sony, as well as Warner Music Group, <\/span>accused Suno of \u201cmassive and ongoing infringement\u201d in a lawsuit that states it \u201ccopied sound recordings from the labels \u2018en masse and ingested them into its <\/span>AI<\/a> model.\u201d<\/span><\/p>\n Music Business Worldwide reported<\/a> that the original complaint listed 560 copyrighted works in a June 2024 filing. As it relates to Suno\u2019s actions, the lawsuit claimed that if there was no copyright violation,<\/span> \u201cSuno\u2019s service would not be able to reproduce the convincing imitations of such a vast range of human musical expression at the quality that Suno touts,\u201d according to<\/a> Forbes.<\/p>\n Suno CEO Mikey Shulman said in a statement at the time that the company \u201cprize(s) originality\u201d and said its technology is \u201cdesigned to generate completely new outputs, not to memorize and regurgitate pre-existing content,\u201d per the outlet.<\/p>\n Warner Music Group ended up settling with Suno in November 2025<\/span> and signed a licensing deal with the company, per Music Business Worldwide<\/span>.<\/span><\/p>\n In the <\/span>latest update, since Suno has reportedly refused to provide details on the material it used to train its AI model, <\/span>UMG and Sony used audio fingerprinting technology<\/a> to search for their recordings in Suno\u2019s training data, according to the outlet. To add their findings to the original complaint, the two record labels filed<\/a> a motion on May 21, 2026,<\/span>\u00a0in the U.S. District Court for the District of Massachusetts.<\/span><\/p>\n \u201cTo identify a sampling of their copied sound recordings for inclusion in an initial Complaint, Plaintiffs engaged in a highly manual and burdensome process. In short, they used \u2018targeted prompts\u2019 in Suno\u2019s service that included the characteristics of popular sound recordings,\u201d the motion read. \u201cIf those targeted prompts caused Suno\u2019s service to generate outputs that closely resembled those particular sound recordings, then Plaintiffs understood that Suno had copied those sound recordings for inclusion in its training data.\u201d<\/p>\n The motion said that Suno acknowledged that \u201cthe tens of millions of recordings that Suno\u2019s model was trained on presumably included recordings whose rights are owned by the Plaintiffs in this case.\u201d<\/p>\n Yet it claims that Suno still refused to identify the sound recordings it used.<\/p>\n