Artificial Intelligence #247

Hey, in this issue: State of AI Report 2024; Pixtral 12B paper; deep cracks in LLMs’ “reasoning” capabilities; hacking generative AI for fun and profit; how symmetry can come to the aid of machine learning; using real-world data to address AI hallucinations; and more.

The sponsors of this issue are Galileo and Particle.

• State of AI Report 2024
• [Google] DataGemma: Using real-world data to address AI hallucinations
• MistralAI released the vision LLM Pixtral 12B paper on ArXiv
• [Sponsored] Learn GenAI best practices from NVIDIA, HP, Databricks, and more

Learn how AI leaders at NVIDIA, HP, Databricks, Twilio, ServiceNow, and more get their GenAI apps in production. Hear how they design a GenAI stack for enterprise scale, including techniques for AI governance, evaluation, and observability. Register for GenAI Productionize 2.0 – free to sign up!

• [Ars Technica] Apple study exposes deep cracks in LLMs’ “reasoning” capabilities
• [Wired] Hacking generative AI for fun and profit
• [Model] Zamba2-7B: An open-weight Apache 2.0 hybrid model composed of Mamba and transformer blocks
• [Sponsored] Hear from some of the biggest names in edge AI and IoT at Particle's Spectra virtual event next week — register now for free!
• [Tool] nanoGPT: The simplest, fastest repository for training/finetuning medium-sized GPTs
• [MIT] How symmetry can come to the aid of machine learning

More than 850,000 subscribers are reading this newsletter. If you are building an AI or a data product or service, you can become a sponsor of one of the future newsletter issues and get your business featured in the newsletter. Feel free to reach out to [email protected] for more details on sponsorships.

Enjoy the newsletter? Please help us make it bigger and better by sharing it with colleagues and friends.