Last Week in AI

#158 - Claude 3, Elon Musk sues OpenAI, StarCoder 2, AI-Generated Spam

03.10.2024 - By Skynet TodayPlay

Download our free app to listen on your phone

Download on the App StoreGet it on Google Play

Our 158th 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 [email protected] and/or [email protected]

Timestamps + links:

Tools & Apps

(00:01:05) Introducing the next generation of Claude

(00:16:05) Competition in AI video generation heats up as Deepmind alums unveil Haiper

(00:19:40) Meta AI creates ahistorical images, like Google Gemini

(00:22:12) Ideogram Is A New AI Image Generator That Obliterates the Competition, Outperforming MidJourney and Dall-E 3

(00:24:25) Wix’s new AI chatbot builds websites in seconds based on prompts

(00:26:34) I used generative AI to turn my story into a comic—and you can too

Applications & Business

(00:27:40) Elon Musk sues OpenAI and CEO Sam Altman for putting profits above humanity , OpenAI Fires Back at Musk Allegations With Trove of Emails

(00:35:42) Inside the Crisis at Google

(00:38:36) It’s official: Waymo robotaxis are now free to use freeways and leave San Francisco

(00:40:33) Nvidia's next-gen AI GPUs could draw an astounding 1000 Watts each, a 40 percent increase — Dell spills the beans on B100 and B200 in its earnings call

(00:42:43) AI chip startup Groq forms new business unit, acquires Definitive Intelligence

Projects & Open Source

(00:45:13) StarCoder 2 and The Stack v2: The Next Generation

(00:49:05) Introducing TripoSR: Fast 3D Object Generation from Single Images

(00:51:21) H2O AI releases Danube, a super-tiny LLM for mobile applications

Research & Advancements

(00:52:27) AtP*: An efficient and scalable method for localizing LLM behaviour to components

(00:58:28) Stable Diffusion 3: Research Paper

(01:02:10) Approaching Human-Level Forecasting with Language Models

(01:05:34) Here Come the AI Worms

(01:08:20) High-speed humanoid feels like a step change in robotics

(01:12:26) Functional Benchmarks for Robust Evaluation of Reasoning Performance, and the Reasoning Gap

Policy & Safety

(01:15:18) India reverses AI stance, requires government approval for model launches

(01:19:16) When Your AIs Deceive You: Challenges with Partial Observability of Human Evaluators in Reward Learning

(01:24:45) OpenAI signs open letter to build AI responsibly just days after Elon Musk sued the company for putting profit ahead of people

(01:27:37) AI-generated articles prompt Wikipedia to downgrade CNET’s reliability rating

(01:29:40) Malicious AI models on Hugging Face backdoor users’ machines

(01:32:25) China offers AI computing ‘vouchers’ to its underpowered start-ups

Synthetic Media & Art

(01:34:18) Trump supporters target black voters with faked AI images

(01:37:10) AI-Generated Kara Swisher Biographies Flood Amazon

(01:39:55) Inside the World of AI TikTok Spammers

(01:43:35) Twitter is becoming a 'ghost town' of bots as AI-generated spam content floods the internet

Fun!

(01:46:35) Man tries to steal driverless car in L.A., doesn’t get far: police

More episodes from Last Week in AI