BROADCASTS.com

Our 142nd episode with a summary and discussion of last week's big AI new.\nApologies for this one coming out after a pause, episodes will resume being released regularly as of this week.\nRead out our text newsletter and comment on the podcast at https://lastweekin.ai/\nEmail us your questions and feedback at contact@lastweekin.ai\nTimestamps + Links:\n(00:00) Intro / Banter\nTools & Apps(03:00) Introducing PlayHT 2.0 Turbo \u26a1\ufe0f - The Fastest Generative AI Text-to-Speech API\n(07:15) YouTube Music now lets you make your own playlist art with AI\n(09:23) Sick of meetings? Microsoft\u2019s new AI assistant will go in your place\n(11:54) Anthropic brings Claude AI to more countries, but still no Canada (for now)\n\n\xa0Applications & Business(14:55) Humanoid robots face a major test with Amazon\u2019s Digit pilots\n(18:40) Figure 01 humanoid takes first public steps\n(22:31) AI-generating music app Riffusion turns viral success into $4M in funding\n(23:35) ChatGPT Creator Partners With Abu Dhabi\u2019s G42 in Middle East AI Push\n(25:00) AMD Scores Two Big Wins: Oracle Opts for MI300X, IBM Asks for FPGAs\n(26:38) Alibaba, Tencent among investors in China\u2019s rival to OpenAI with $341 million funding\n(30:35) AI companies drive demand for office space in tech hubs, new study finds\n(32:13) OpenAI is in talks to sell shares at an $86 billion valuation\n\nProjects & Open Source(35:00) Introducing Video-To-Text and Pegasus-1 (80B)\n(39:35) Adept Releases Fuyu-8B for Multimodal AI Agents\n(42:03) MiniGPT-v2: large language model as a unified interface for vision-language multi-task learning\n(44:53) Meta\u2019s Habitat 3.0 simulates real-world environments for intelligent AI robot training\n(48:22) DeepMind UniSim simulates reality to train robots, game characters\xa0\n(49:13) Jina AI Launches World's First Open-Source 8K Text Embedding, Rivaling OpenAI\n(51:13) Llemma: An Open Language Model For Mathematics\n\nResearch & Advancements(53:22) Eliciting Human Preferences with Language Models\n(57:23) New Nvidia AI agent, powered by GPT-4, can train robots\n(01:01:38) Unveiling the General Intelligence Factor in Language Models: A Psychometric Approach\n(01:04:48) AgentTuning: Enabling Generalized Agent Abilities for LLMs\n(01:09:51) Contrastive Prefence Learning: Learning from Human Feedback without RL\n(01:11:25) \u2018Mind-blowing\u2019 IBM chip speeds up AI\xa0\n\nPolicy & Safety(01:14:57)\xa0 GM Cruise unit suspends all driverless operations after California ban\n(01:18:52) AI researchers uncover ethical, legal risks to using popular data sets\n(01:22:22) AI Safety Summit: day 1 and 2 programme\n(01:25:23) Anthropic's AI chatbot Claude is posting lyrics to popular songs, lawsuit claims\n(01:26:38) Mike Huckabee says Microsoft and Meta stole his books to train AI\n(01:27:10) Clearview AI Successfully Appeals $9 Million Fine in the U.K.\n(01:28:11) North Korea experiments with AI in cyber warfare: US official\n(01:30:17) OpenAI forms new team to assess \u2018catastrophic risks\u2019 of AI\nUK poised to establish global advisory group on AI\n\nSynthetic Media & Art(01:32:22) This new data poisoning tool lets artists fight back against generative AI\n(01:34:32) Amazon now lets advertisers use generative AI to pretty up their product shots\n(01:36:36) The Beatles: \u2018final\u2019 song Now and Then to be released thanks to AI technology