- AI KATANA
- Posts
- Apple, Nvidia, Anthropic Used Thousands of Swiped YouTube Videos to Train AI
Apple, Nvidia, Anthropic Used Thousands of Swiped YouTube Videos to Train AI
Also: AI budgets poised to surge in 2025
Check out our new AI jobs section below, we will be featuring AI jobs in Asia. If you would like to feature a role, please contact us.
Morning!
In today’s newsletter, a Proof News investigation exposes Apple, Nvidia, and Anthropic for using transcripts from 173,536 YouTube videos to train AI models without creators’ consent, raising ethical and legal concerns. Meanwhile, the South Korean beauty industry is leveraging AI to develop personalized skincare and makeup products, accelerating market growth. NASA’s Mars rover, Perseverance, is utilizing AI to enhance autonomous scientific operations on the Red Planet. U.S. companies plan to nearly double their AI investments by 2025, though many lack the infrastructure and talent to maximize AI’s potential. In AI tools updates, Anthropic’s Claude is now available on Android, offering advanced generative AI capabilities. Finally, UK AI startups have reached a valuation of $256 billion, leading Europe in venture capital investment.
Sliced:
⚠️ Apple, Nvidia, Anthropic Used Thousands of Swiped YouTube Videos to Train AI
🇰🇷 South Korean Beauty Industry Using AI to Develop New Products
👾 Here’s How AI Is Changing NASA’s Mars Rover Science
📈 AI budgets poised to surge in 2025
A Proof News investigation revealed that major AI companies, including Apple, Nvidia, and Anthropic, used transcripts from 173,536 YouTube videos to train AI models without creators’ knowledge or consent. This dataset, called YouTube Subtitles, included educational content from institutions like Harvard and MIT, as well as popular shows and YouTube personalities. Despite YouTube’s terms prohibiting automated scraping, the subtitles were collected and incorporated into a larger repository known as the Pile, used extensively by tech giants for developing AI technologies. The discovery has sparked concerns about the ethical and legal implications of using creators’ work without permission, as well as the potential for AI to replicate and replace human content creation.
The South Korean beauty industry is leveraging AI to innovate and develop new products, enhancing personalization and customer satisfaction. Companies like AmorePacific are using AI technologies in their laboratories to create customized skincare and makeup products. AI systems analyze skin tone and condition to recommend personalized solutions, such as foundation and lipstick shades, which robots then mix on-site. This approach has led to faster product development and increased accuracy, as AI reduces human error. The global beauty market, which reached $625.6 billion in 2023, is expected to see further growth with AI-driven advancements, predicting a market expansion to $8.1 billion by 2028.
AI is revolutionizing NASA’s Mars rover science by enabling more efficient and autonomous operations. The Perseverance rover utilizes AI to analyze rock compositions in real-time, marking the first use of such technology on Mars. The AI-powered Planetary Instrument for X-ray Lithochemistry (PIXL) autonomously positions itself to scan rocks, determining their chemical makeup to assess past conditions for microbial life. This adaptive sampling reduces the need for human intervention, allowing PIXL to make immediate decisions and gather more data efficiently. AI also aids the rover’s navigation and laser-based rock analysis, enhancing mission productivity and scientific discovery on the Red Planet.
U.S. companies are set to nearly double their investments in AI, with 30% planning to spend at least $10 million in 2025, up from 16% currently, according to an Ernst & Young survey. Despite this surge, many firms are neglecting the necessary infrastructure and talent development required to fully leverage AI’s potential. The report highlights the transition of AI from a hype phase to a productivity tool, emphasizing the need for a comprehensive strategy to integrate AI across enterprises. Success with AI hinges on robust data infrastructure and a skilled workforce, without which investments may falter. The growing focus on ROI for AI investments is reflected in similar findings from KPMG, showing a shift towards maximizing revenue and competitive advantage through generative AI.
🧑🏽💻 AI Jobs
🛠️ AI tools updates
Anthropic has launched an Android version of its Claude app, following its debut on iOS two months earlier. Claude is a generative AI tool capable of handling prompts involving text, files, and images, performing tasks like drafting, summarizing, and answering complex questions. The app is based on the Claude 3 LLM family of AI models and offers a simple user interface similar to its iOS counterpart. While the app is free, a Pro plan is available for increased usage and access to advanced models. Claude aims to provide reliable and accurate assistance, emphasizing Anthropic’s commitment to creating safe and dependable AI tools.
💵 Venture Capital updates
UK AI startups have reached a collective valuation of $256 billion, now accounting for 22% of the country’s innovation ecosystem, up from 12% in 2019. In the first half of 2024, these startups raised $2.1 billion, with projections indicating this amount could more than double in the next five months. London-based Wayve, specializing in AI for autonomous vehicles, secured the largest investment with a $1.1 billion Series C round. The UK leads Europe in venture capital investment, raising $9.4 billion in H1 2024, significantly higher than France and Germany, which each raised $4.3 billion. AI investment in Europe has also grown, now constituting 18% of the continent’s VC funding.
🫡 Meme of the day
⭐️ Generative AI image of the day
Before you go, check out The secrets to keeping machine learning engineers moving fast.