Sony Music Group‘s parent company, Japan-headquartered Sony Group, has reportedly developed technology to identify copyrighted music embedded in AI-generated tracks. That’s according to the financial news outlet Nikkei Asia, which reported on Monday (February 16) that the tech has opened a path for songwriters to claim compensation when their work is used without authorization.
Source: Sony Group develops tech to track original music in AI-generated songs (report)
The Motion Picture Association on Thursday denounced the newest AI video generator, Seedance 2.0, saying it had unleashed a flood of copyright infringement in the day since it became available. ByteDance, the Chinese owner of TikTok, said the model represents a “substantial leap in generation quality” since the previous version.
Amazon joins Microsoft, which recently launched Publisher Content Marketplace, in developing platforms to allow publishers to license content and be paid-per-use, rather than by offering an blanket, up-front fee. The two tech giants are not operating in a vacuum, as a variety of companies have been racing to establish their their own licensing platforms.
If enacted as drafted, the bill would establish mandatory reporting requirements for companies developing artificial intelligence (AI) models that are trained using original works that are protected under U.S. copyright law, and would create an additional cause of action for copyright owners alleging that generative AI developers failed to give such notice with respect to their works.
Source: CLEAR Act Would Establish Notice Requirements for Copyrighted Works in AI Training Data
LightBar, founded by an anonymous CEO with credentials in the fintech world, wants to become a gun for hire for studios, enlisting internet users to hunt down AI models generating content that could have been trained on copyrighted material. Or, to put it another way, LightBar is assembling bounty hunters to track down misuse of Hollywood IP.
Source: Big AI Beware: Bounty Hunters Are Coming To Track Down Misuse Of Hollywood IP
Despite being sued by Spotify and several record labels, Anna’s Archive has silently begun releasing the actual music files from its massive Spotify scrape. The shadow library’s backend torrent index now lists dozens of new torrents containing approximately 2.8 million tracks totaling roughly 6 terabytes of audio data. This marks a significant escalation in the already historic standoff with the music industry.
Source: Anna’s Archive Quietly ‘Releases’ Millions of Spotify Tracks, Despite Legal Pushback * TorrentFreak
Spotify says it wants to enable listeners to interact with their favorite artists’ music through AI – and that the technology to do so is already built. The barrier? Licensing. The comments land amid an intensifying industry debate over how AI-powered derivatives of existing music should be licensed and monetized – and where fans should be allowed to create them.
Performance rights collection organization SoundExchange has added 17 new agreements with collective management organizations worldwide in recent months. The expansion includes new partnerships in Kenya, Barbados, Panama, Paraguay, and several European territories, bringing SoundExchange’s total portfolio to over 90 agreements with international CMOs
Swedish music streaming giant Spotify saw its user numbers peak last quarter, driven by its year-end “Wrapped” campaign, which rounds up stats and listening highlights for users and new features on its free tier. Spotify said the “Wrapped” campaign resulted in more than 300 million engaged users and 630 million shares on social media in 56 languages.
Source: Spotify hits a record 751M monthly users thanks to Wrapped, new free features | TechCrunch
Blockchain and centralized platforms promise to address trust in visual content through commercial gatekeeping. These approaches share a fatal flaw that has nothing to do with their technical elegance: they gate trust behind commercial access. And in doing so, they create systems that, by their very design, cannot serve the function that trust infrastructure requires.
Source: Why Trust for Visual Content Cannot Be Solved by Centralized Commercial Solutions – Kaptur