DeepSummary
The hosts, Andrei Kurenkov and guest co-host Jon Krohn, discuss the latest developments in artificial intelligence, including new AI music generators, Microsoft's efforts to compete with GPT-4, DeepMind's AlphaFold3 for modeling DNA, and a proposed xLSTM architecture as an alternative to transformers. They also cover OpenAI's release of a set of guidelines called Model Spec, aimed at outlining principles for AI behavior.
The episode delves into various open-source projects, research papers, and advancements in the field of AI, such as tools for evaluating language models, code intelligence models, and multimodal evaluation suites. Discussions also touch on policy and safety concerns, including potential export controls on AI models and the use of AI-equipped robot dogs for military purposes.
The hosts explore the growing presence of AI-generated media, including Audible's adoption of AI-narrated audiobooks, the labeling of AI-generated content on TikTok, and the emergence of deepfakes that can deceive people, including in romance scams. They highlight examples of AI-generated images that have gone viral, fooling people into believing they were real.
Key Episodes Takeaways
- New AI music generators, such as ElevenLabs, are pushing the boundaries of AI-generated audio content, potentially disrupting the music industry.
- Microsoft is developing its own large language model, codenamed Mai 1, to compete with OpenAI's GPT-4 and Google's Gemini, signaling increased competition among tech giants in the AI race.
- DeepMind's AlphaFold3 expands the capabilities of the AlphaFold system to model DNA and other biomolecular interactions, advancing the application of AI in scientific research.
- Researchers are exploring alternative architectures to transformers, such as xLSTM, aiming to develop more efficient and scalable models while retaining or improving performance.
- OpenAI's Model Spec outlines guidelines and principles for AI behavior, reflecting the growing need for ethical frameworks and governance as AI systems become more capable and influential.
- The integration of AI into various industries, such as audiobook narration and content creation on platforms like TikTok, is rapidly increasing, raising concerns about the impact on human creators and the need for labeling AI-generated content.
- The prevalence of deepfakes and AI-generated media is increasing, leading to concerns about misinformation, scams, and the erosion of trust in visual and audio content.
- Policy discussions around AI are intensifying, with proposals for export controls on AI models and the development of AI-equipped military systems, highlighting the need for responsible governance and oversight.
Top Episodes Quotes
- “But you know, there's so many labs trying to do this kind of thing. And it is so important in order to make AI be something that is broadly accessible and not super expensive, not requiring more and more and more nuclear power plants to be set up to run AI centers.“ by Jon Krohn
- “It's interesting to me that the thing that got the biggest chuckle from me on your episode last week was you saying, andre, that China, the country, did not release this model. You see that so often in headlines of, like, China releases.“ by Jon Krohn
- “Yeah, I think as we've spoken about this for 30 seconds more, my mind has kind of shaken off some cobwebs and reminded me that I think when I first heard about it, maybe a year ago, I did do some searches and I was like, wow, it's really nice Ui, but it didn't scratch any actual informational need for me beyond what I could already retrieve very conveniently with either a Google search or a generative AI search with one of the tools that I was already using.“ by Jon Krohn
Entities
Company
Person
Product
Book
Episode Information
Last Week in AI
Skynet Today
5/12/24
Our 166th episode with a summary and discussion of last week's big AI news!
Read out our text newsletter and comment on the podcast at https://lastweekin.ai/
Email us your questions and feedback at contact@lastweekin.ai and/or hello@gladstone.ai
Timestamps + links:
- (00:00:00) Intro / Banter
- Tools & Apps
- (00:04:23) ElevenLabs previews music-generating AI model
- (00:09:31) Mysterious “gpt2-chatbot” AI model appears suddenly, confuses experts
- (00:13:00) SoundHound AI and Perplexity Partner to Bring Online LLMs to Next Gen Voice Assistants Across Cars and IoT Devices
- (00:14:50) Stability AI sows gen AI discord with Stable Artisan
- (00:16:35) Apple Will Revamp Siri to Catch Up to Its Chatbot Competitors
- (00:18:54) Alibaba rolls out latest version of its large language model to meet robust AI demand
- Applications & Business
- (00:19:34) OpenAI and Stack Overflow partner to bring more technical knowledge into ChatGPT
- (00:17:31) New Microsoft AI model may challenge GPT-4 and Google Gemini
- (00:31:08) Wayve, an A.I. Start-Up for Autonomous Driving, Raises $1 Billion
- (00:32:00) Motional delays commercial robotaxi plans amid restructuring
- (00:33:54) The rise of the Chinese AI unicorns doing battle with OpenAI
- Projects & Open Source
- (00:35:25) Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models
- (00:40:12) DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
- (00:44:31) OpenVoice V2: Evolving Multilingual Voice Cloning with Enhanced Style Control and Cross-Lingual Capabilities
- (00:45:20) Granite Code Models: A Family of Open Foundation Models for Code Intelligence
- (00:46:00) Hugging Face launches LeRobot open source robotics code library
- (00:48:50) Vibe-Eval: A new open and hard evaluation suite for measuring progress of multimodal language models
- Research & Advancements
- (00:50:02) Google DeepMind’s Groundbreaking AI for Protein Structure Can Now Model DNA
- (00:57:20) xLSTM: Extended Long Short-Term Memory
- (01:06:35) StoryDiffusion: Consistent Self-Attention for Long-Range Image and Video Generation
- (01:07:55) Chain of Thought Empowers Transformers to Solve Inherently Serial Problems
- (01:11:48) KAN: Kolmogorov-Arnold Networks
- Policy & Safety
- (01:13:20) US lawmakers unveil bill to make it easier to restrict exports of AI models
- (01:17:30) OpenAI’s Model Spec outlines some basic rules for AI
- (01:20:18) Robot dogs armed with AI-targeting rifles undergo US Marines Special Ops evaluation
- (01:25:15) OpenAI Releases ‘Deepfake’ Detector to Disinformation Researchers
- Synthetic Media & Art
- (01:28:15) Audible’s Test of AI-Voiced Audiobooks Tops 40,000 Titles
- (01:32:30) TikTok will automatically label AI-generated content created on platforms like DALL·E 3
- (01:33:23) Katy Perry's Fan-Made AI Image Is So Real It Fooled the World Into Thinking She Was at the Met Gala
- (01:35:32) South Korean woman falls for deepfake Elon Musk, loses $50K in romance scam
- (01:37:18) Why young Russian women appear so eager to marry Chinese men
- (01:40:18) AI Outro Song