Épisodes

  • NVIDIA's Jetson Orin Nano, FACTS Benchmark, YouTube’s AI Celeb tools, & Falcon 3’s LLM
    Dec 18 2024

    In today’s AI Deep Dive, we unpack NVIDIA's Jetson Orin Nano Super, a game-changer for affordable generative AI. Explore the FACTS Grounding Leaderboard, a new benchmark assessing large language models' factual accuracy. We also discuss YouTube's collaboration with Creative Artists Agency to safeguard celebrity identities in AI-generated content. Finally, meet Falcon 3, an open-source LLM delivering top-tier performance on lightweight devices, redefining AI accessibility and innovation.

    Voir plus Voir moins
    8 min
  • Google's Veo 2 & Imagen 3, ChatGPT's Search Expansion, and Meta Glasses Latest Updates
    Dec 17 2024

    In this episode of AI Deep Dive, we explore Google's groundbreaking Veo 2 and Imagen 3, setting new standards in AI video and image generation. OpenAI expands its ChatGPT Search tool to all users, addressing accessibility and sparking publisher debates. Meta updates its Ray-Ban Meta glasses with cutting-edge AI features, including live translation and music identification. Finally, YouTube empowers creators with new settings to control AI training access, tackling unauthorized use of their content.

    Voir plus Voir moins
    11 min
  • Explore Anthropic's Clio, Amazon's AI Topics, and Airbnb's Anti-Party AI Tech
    Dec 16 2024

    In this episode of AI Deep Dive, we explore Anthropic’s Clio, a privacy-preserving tool offering new insights into AI usage while improving safety. Amazon Prime Video's AI Topics feature personalises streaming recommendations like never before. Meanwhile, Airbnb’s global anti-party AI technology prevents disruptive bookings, reshaping the hospitality industry. Lastly, Sakana AI's universal transformer memory sets a new standard in LLM optimisation, cutting costs by 75%.

    Voir plus Voir moins
    11 min
  • Meta vs. OpenAI, ByteDance Bets on AI Video, and AI World Models
    Dec 15 2024

    In this episode of AI Deep Dive, we unpack Meta’s opposition to OpenAI’s for-profit model and explore ByteDance’s pivot toward AI video generation over chatbots. Discover the transformative potential of AI world models in generating realistic videos and simulating complex scenarios, alongside the challenges they face. We also examine Klarna CEO’s bold claims about AI replacing human workers, contrasting his statements with the company’s ongoing hiring practices.

    Voir plus Voir moins
    13 min
  • Grok Updates, Character AI Teen Safety Features, and Superintelligent AI
    Dec 14 2024

    In today’s AI Deep Dive, we cover Elon Musk’s xAI releasing a faster Grok chatbot with new API features and a "Grok button" for X. Character AI responds to lawsuits by implementing teen safety features and rebranding as an entertainment platform. We also discuss OpenAI co-founder Ilya Sutskever’s views on superintelligent AI and reflect on the legacy of Suchir Balaji, a former OpenAI employee who raised critical copyright concerns before his tragic passing.

    Voir plus Voir moins
    9 min
  • OpenAI’s Video Chat, Microsoft’s Phi-4, Claude 3.5 Haiku, & Meta’s Video Seal
    Dec 13 2024

    In today’s AI Deep Dive, we explore OpenAI’s real-time video integration for ChatGPT and Microsoft’s Phi-4 model revolutionizing math problem-solving. We also delve into Anthropic’s Claude 3.5 Haiku, enhancing conversational AI, and Meta’s Video Seal, an open-source tool to watermark AI-generated videos and combat deepfakes. Join us for a comprehensive look at the latest AI advancements and their implications.

    Voir plus Voir moins
    9 min
  • Apple Intelligence, Google’s Astra and Mariner Projects, OpenAI Outage, & BRICS AI Alliance
    Dec 12 2024

    In today’s AI Deep Dive, we unpack Apple’s major AI updates, including Genmoji and Image Playground, plus the global expansion of its Writing Tools. We also explore Google’s innovative Project Astra and Project Mariner, multimodal AI and browser-based agents designed for complex tasks. OpenAI’s simultaneous outage raises questions, and Russia’s push for a BRICS AI alliance signals geopolitical shifts in the global AI race.

    Voir plus Voir moins
    15 min
  • Google’s Gemini 2.0, Replit’s AI Assistant, & Target's Bullseye Gift Finder
    Dec 11 2024

    AI innovation is taking center stage as Google unveils Gemini 2.0, a groundbreaking AI model designed for image and audio generation, and to power Google products and AI agents. Replit enters the AI space with its Agent and Assistant tools, streamlining software creation with natural language interactions. Target introduces Bullseye Gift Finder and Shopping Assistant, revolutionizing online shopping through AI-powered personalization. Meanwhile, WaveForms, an AI startup founded by former OpenAI talent Alexis Conneau, is developing emotionally intelligent audio AI to enhance conversations. Join us in this episode of AI Deep Dive as we explore these transformative developments in AI!

    Voir plus Voir moins
    9 min