Several developments in the AI landscape are unfolding across various sectors. Meta is considering a significant investment of over $10 billion in Scale AI, a company specializing in data infrastructure for AI model training, including Meta's own Llama. This move underscores Meta's commitment to expanding its AI capabilities, with CEO Mark Zuckerberg planning to invest up to $65 billion in AI this year. Meanwhile, Apple is testing AI-powered app tagging in its iOS 26 developer beta to enhance app discoverability by analyzing screenshots and descriptions. In the UK, the Data Bill has been passed, allowing AI developers to use copyrighted material for training with an opt-out option for rights holders, though AI copyright issues remain a topic for future legislation. Musician Timbaland faces accusations of using stolen beats to train his AI Suno platform, sparking debate over consent and ethics in AI music creation. The UN warns that AI is rapidly approaching human-like intelligence (AGI) and calls for global coordination to manage associated risks, suggesting a special General Assembly and a global AGI observatory. The Venice Biennale is experimenting with AI-generated summaries in project descriptions to promote accessibility and critical thinking about AI. Concerns are also rising about the potential displacement of American jobs by AI, with calls for President Trump to protect workers and tax AI transactions that replace human labor. Multimodal AI, which processes various data formats, offers significant power but faces challenges related to data integration, bias, security, and privacy. Finally, experts are warning that the U.S. could lose its lead in AI to China, with Chinese companies narrowing the gap, prompting recommendations for Washington to make its AI models more accessible and efficient.
Key Takeaways
- Meta is considering investing over $10 billion in Scale AI to bolster its AI capabilities.
- Apple's iOS 26 beta is testing AI-powered app tagging to improve app discoverability.
- The UK Data Bill allows AI developers to use copyrighted material for training, with an opt-out option for rights holders.
- Timbaland is accused of using stolen beats to train his AI Suno platform, raising ethical questions.
- The UN warns AI is nearing human-like intelligence (AGI) and calls for global coordination.
- The Venice Biennale is experimenting with AI-generated summaries to promote critical thinking about AI.
- There are growing concerns about AI potentially displacing American jobs.
- Multimodal AI offers power but faces data integration, bias, security, and privacy challenges.
- Experts warn the U.S. could lose its lead in AI to China.
- Scale AI expects to generate $2 billion in revenue this year, doubling last year's earnings.
Meta may invest $10B in AI firm Scale AI
Meta is considering investing over $10 billion in Scale AI, which could be one of the largest private AI deals. Scale AI provides labeled data for training AI models like ChatGPT and Meta's Llama. Meta previously invested in Scale's Series F round in 2024, valuing the company at $18 billion. Scale AI expects to generate $2 billion in revenue this year, doubling last year's earnings. Scale AI CEO Alexandr Wang advises lawmakers on AI regulation, and the company contributes to Meta's Defense Llama model for U.S. national security.
Meta may invest $10B in AI firm Scale AI
Meta is considering investing over $10 billion in Scale AI, which could be one of the largest private AI deals. Scale AI provides labeled data for training AI models like ChatGPT and Meta's Llama. Meta previously invested in Scale's Series F round in 2024, valuing the company at $18 billion. Scale AI expects to generate $2 billion in revenue this year, doubling last year's earnings. Scale AI CEO Alexandr Wang advises lawmakers on AI regulation, and the company contributes to Meta's Defense Llama model for U.S. national security.
Meta considers $10B investment in AI data firm Scale AI
Meta is in talks to invest over $10 billion in Scale AI, a company specializing in data infrastructure for artificial intelligence. Scale AI supports large language models and works with Meta on 'Defense Llama'. The potential investment would be Meta’s largest external AI commitment. Meta CEO Mark Zuckerberg plans to invest up to $65 billion in AI this year. This deal shows Meta's focus on expanding its AI capabilities through partnerships.
Apple's iOS 26 beta uses AI to tag App Store apps
Apple is testing AI-powered app tagging in the iOS 26 developer beta to improve app discoverability. The AI analyzes screenshots and descriptions to create tags for better classification and ranking. Developers can see and control the AI-generated metadata associated with their apps. The new system considers screenshot captions and in-app user interface text for app ranking. Developers should ensure app visuals and descriptions accurately reflect app features to benefit from the AI tagging system.
The AI Industry Is Ready to Get Rich off Trump\u2019s Defense Department
Oops. This link is broken, but the democratic experiment endures. Find more ennobling content back at ...
UK passes Data Bill, AI copyright battle may continue
The UK Data Bill has passed after being defeated five times in the House of Lords. An amendment requiring AI developers to disclose data used in training models was removed. The bill allows AI developers to use copyrighted material for training without permission but gives rights holders the option to opt-out. The government plans to address copyright and AI questions in a separate AI bill. Supporters fear aligning with the Trump administration could weaken protections for creators.
Timbaland accused of using stolen beat for AI training
Timbaland is facing criticism for allegedly using producer K Fresh Music's beat and another creator's verse to train his AI Suno platform. He is accused of taking the producer tag without permission. Some see this as remixing, while others condemn using music for AI without consent. Timbaland's original video showcasing the AI remixing tool dates back to October 2024. Some defend Timbaland's involvement in AI, emphasizing the importance of diversity in the field.
UN warns AI is nearing human intelligence, urges action
The United Nations warns that AI is rapidly approaching human-like intelligence, called Artificial General Intelligence (AGI). A UN report calls for global coordination to manage the risks of AGI. The report says AGI could accelerate scientific discoveries but also poses risks like autonomous harmful actions. The UN suggests a special General Assembly on AGI and a global AGI observatory. DeepMind CEO Demis Hassabis suggests a UN-like organization to oversee AGI development.
AI-generated wall texts at Venice Biennale spark debate
The Venice Biennale used AI-generated summaries in project descriptions as an experiment. The summaries juxtapose human intelligence with artificial intelligence. The AI-generated summaries were not edited to show both the benefits and shortcomings of AI. The goal is to promote accessibility and critical thinking about AI in curatorial contexts. The experiment aims to normalize a culture of accessibility within architecture exhibitions.
Trump urged to protect American jobs from AI
An opinion piece urges President Trump to protect American workers from job losses due to AI. Millions of jobs could be eliminated by AI because it is cheaper and quieter. The author argues Trump should defend American workers and not let Silicon Valley control the economy. Companies taking federal money should not be allowed to replace workers with AI. The author suggests taxing AI transactions that replace human jobs to fund job retraining programs.
Multimodal AI offers power, faces data and bias challenges
Multimodal AI processes information across different formats like text, images, audio, and video. It helps businesses operate, innovate, and compete by unifying data sources. It can improve customer support, predict equipment failures, and personalize experiences. However, implementing multimodal AI is complex due to data integration and potential biases. Organizations need to integrate data flows and address biases in visual and language datasets. Multimodal systems also raise concerns about data security and privacy.
Experts warn US could lose AI race to China
Experts suggest the U.S. may lose its lead in artificial intelligence to China. Chinese companies like DeepSeek and Huawei are narrowing the gap. Analysts recommend Washington make its AI models more open and appealing. They suggest focusing on building cheaper, powerful, and efficient models for global markets.
Sources
- Meta In Talks To Invest Over $10B In Scale AI — Possibly The Biggest Private AI Deal This Year - Microsoft (NASDAQ:MSFT), Meta Platforms (NASDAQ:META)
- Meta In Talks To Invest Over $10B In Scale AI — Possibly The Biggest Private AI Deal This Year
- Meta (META) Eyes $10 Billion Investment in Scale AI to Accelerate AI Strategy
- Apple Brings AI-Powered App Store Tagging to iOS 26 Developer Beta: What It Means for App Rankings
- The AI Industry Is Ready to Get Rich off Trump’s Defense Department
- UK Data Bill (Finally) Passes, But Battle Over Copyright and AI Transparency Could Continue
- Timbland Accused Of Stealing Producer's Beat For A.I. Training
- UN Sounds Alarm On AI Nearing Human-Like Intelligence 'AGI', Urges Action
- The Real Point of Using A.I.-Generated Wall Texts at the Venice Biennale
- Trump must stop the AI bloodbath before it’s too late
- Multimodal AI: A Powerful Leap With Complex Trade-Offs
- US should prepare for possibility of losing AI race to China, experts say