Major music labels are in negotiations with AI music startups Suno and Udio to address copyright concerns and establish licensing agreements for the use of artists' music in AI training. These discussions aim to compensate artists and labels when AI utilizes their music, implement systems for identifying copyrighted material in AI-generated content, and potentially involve the labels in the commercial application of AI music tools, possibly through ownership stakes. Simultaneously, the increasing energy consumption of AI is raising concerns, with AI hardware and data centers potentially using a significant portion of global electricity by 2025 and 2030, respectively. Experts emphasize the importance of understanding and mitigating biases in AI data through education, while initiatives like LawZero are being launched to ensure AI's safe and beneficial development. Guidelines are also being developed for the safe use of AI in specialized fields like medicine, and there's a growing awareness of AI's limitations as people become more critical of AI's attempts to mimic human interaction.
Key Takeaways
- Major music labels are negotiating with AI music startups Suno and Udio to settle copyright lawsuits and establish licensing agreements.
- The music industry aims to ensure artists are compensated when their music is used to train AI models.
- AI's energy consumption is rapidly increasing, with data centers potentially using a significant portion of global electricity in the coming years.
- Measuring AI's energy use is complex, and experts suggest focusing on the energy sources used by AI companies.
- It's crucial to teach students how to identify and address biases in AI training data.
- Yoshua Bengio is launching LawZero, a nonprofit focused on making AI safer for humanity.
- New guidelines are being developed for the safe and ethical use of AI in medicine.
- Growing public annoyance with AI's attempts to mimic human interaction may indicate AI will remain a tool rather than a competitor.
- AI hardware startups are gaining traction due to infrastructure needs.
Music labels may settle AI lawsuits with Suno and Udio
Major music labels like Universal, Warner, and Sony are discussing deals with AI music startups Suno and Udio. These deals could settle copyright lawsuits over how AI uses music. The labels want fees and ownership in the companies for letting them use music to train AI. This could set a standard for how AI companies pay artists and music companies.
Music industry negotiates AI music licensing with Suno and Udio
Music giants like Sony, Warner, and Universal are talking with AI companies Suno and Udio about licensing music. They want to make sure artists get paid when their music trains AI models. The talks come as Suno and Udio face a lawsuit for using copyrighted music. Labels want technology to track when AI uses their music and want a say in new AI music products.
Music labels seek deals with AI startups Suno and Udio
Major music labels are in talks with AI startups Suno and Udio for licensing agreements. The labels want to be paid when AI uses their artists' music, like Taylor Swift and Drake. They also want systems to identify copyrighted material in AI-generated music. The labels want input on how AI music tools are used commercially and may take ownership stakes in Suno and Udio.
AI's energy use is surprisingly complex and growing fast
AI uses a lot of energy because it needs powerful computers and data centers. These data centers store huge amounts of information and run complex calculations. AI's energy use is growing as more people use AI tools. Data centers may use 7.5% of all US electricity by 2030. Experts are still trying to figure out exactly how much energy AI uses.
Measuring AI's energy use is harder than you think
Calculating AI's energy use is difficult, but important for understanding its impact. AI companies are developing more complex models that use even more energy. AI video creation is especially energy-intensive. Experts say that focusing on how AI companies get their energy is more important than worrying about individual AI use.
Will Meta AI take our jobs?
This article is for PRWeek subscribers only.
AI hardware uses a lot of energy
AI hardware needs a lot of energy to train on data and answer questions. In 2025, AI hardware could use as much electricity as countries like Switzerland or Austria. This could be 11% to 20% of the energy used by all data centers.
Teach students to spot bias in AI data
AI models can have problems if they are trained on data that is not diverse. For example, models trained mostly on data from white males may not work well for other groups. It's important to teach students how to find and fix these biases in AI data. Experts say that understanding the data is just as important as building the AI model.
AI hardware startups see opportunity in infrastructure needs
This article discusses the increasing interest in AI hardware startups due to infrastructure bottlenecks.
Scientist plans LawZero to make AI safer for humanity
Yoshua Bengio, a top computer scientist, is starting a nonprofit called LawZero to make AI safer. LawZero aims to avoid profit motives and involve governments in AI's rules. Bengio wants to ensure AI benefits humanity and avoids harm.
AI in medicine: 8 ways for companies to use it safely
A new report gives advice on using AI safely in medicine. It helps companies follow rules like the EU AI Act. The report says companies should manage risks, have people oversee AI, and check AI's work. The FDA in the U.S. also has guidelines for using AI in medicine.
Are we getting annoyed with AI? That's a good sign
People are starting to get annoyed with AI's attempts to act like a human. AI chatbots are trying too hard to be friendly and helpful. This may be a good thing because it shows that AI is still not a real person. AI may remain a tool instead of a competitor.
Ray Dalio says AI is starting a new era
...
Sources
- Major Labels in Licensing Talks With A.I. Companies Suno & Udio Amid Blockbuster Lawsuit
- Music giants begin negotiating AI licensing rights for labels and artists
- AI Startups: Music Labels Push for Licensing Deals With Suno, Udio
- How much energy does AI really use? The answer is surprising - and a little complicated
- Inside the tedious effort to tally AI’s energy appetite
- Is Meta AI coming for our jobs?
- AI energy usage, rough estimates for the hardware
- Q&A: How to help students recognize potential bias in their AI datasets
- AI infrastructure bottlenecks drive hardware startup interest
- The Most-Cited Computer Scientist Plans to Make AI More Trustworthy
- Artificial Intelligence in Pharmacovigilance: Eight Action Items for Life Sciences Companies | Data Matters Privacy Blog
- Our Annoyance With AI Might Be a Good Sign
- Ray Dalio on AI: 'We are now at the brink of a new era'