Artificial Intelligence #235

Hey, in this issue: Llama 3.1, the most capable open-weight models; FlashAttention-3; AI-based climate modeling; AI is taking jobs in the video game industry; data flywheels for LLM applications; creating and verifying stable AI-controlled systems; and more.

The sponsors of this issue are Nexla and True Positive Inc.

• [Meta] Introducing Llama 3.1: Our most capable models to date
• FlashAttention-3: Fast and accurate attention with asynchrony and low-precision
• [Google] Fast, accurate climate modeling with NeuralGCM
• [Sponsored] Amazon Bedrock Tech Talk: Data Integration Using Nexla

Streamline & scale data integration to and from Amazon Bedrock for GenAI applications. Feat. Sathya B. (Senior Solutions Architect at AWS): Feed unstructured data into Amazon Bedrock | Turn text & documents into vector embeddings | Apply Nexla’s RAG data flow capabilities to enhance AI deployment.

• [Wired] AI is already taking jobs in the video game industry
• Data flywheels for LLM applications
• [MIT] Creating and verifying stable AI-controlled systems in a rigorous and flexible way
• [Sponsored] Machine Learning Engineering book
• Researchers are figuring out how large language models work
• [MIT] How to assess a general-purpose AI model’s reliability before it’s deployed

More than 800,000 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.