Home

Think

Podcasts

Mixture of Experts

Episode 44

Claude 3.7 Sonnet, BeeAI agents, Granite 3.2 and emergent misalignment

Watch the episode
Episode 44: Claude 3.7 Sonnet, BeeAI agents, Granite 3.2 and emergent misalignment

IBM® Granite™ 3.2 is officially here! In episode 44 of Mixture of Experts, join host Tim Hwang and experts Kate Soule, Maya Murad and Kaoutar El Maghraoui to debrief a few big AI announcements. Last week, we covered small vision-language models (VLMs), and this week Granite 3.2 dropped with new VLMs, enhanced reasoning capabilities and more! Join Kate as she takes us under the hood to understand the new features and how they were created. 

Anthropic dropped a new intelligence model, Claude 3.7 Sonnet, and a new agentic coding tool, Claude Code. Hear the experts explore: why did Anthropic release these separately? Then, as we cannot have an episode without covering agents, Maya takes us through the new BeeAI agents! Finally, can fine-tuning on a malicious task lead to much broader misalignment? Our experts analyze a new paper released on ‘Emergent misalignment’ to uncover the risks. All this and more on this week's episode!

Key takeaways:

  • 00:01 – Intro
  • 00:35 – Claude 3.7 Sonnet
  • 11:58 – BeeAI agents 
  • 22:17 – Granite 3.2
  • 32:31 – Emergent misalignment

The opinions expressed in this podcast are solely those of the participants and do not necessarily reflect the views of IBM or any other organization or entity.

View all Mixture of Experts episodes
Listen on Apple Podcasts Spotify Podcasts Casted YouTube
Explore more episodes Deep Research, OpenAI inference chip, small VLMs and AI agent job posting

What is the hype with Deep Research? In episode 43 of Mixture of Experts, we cover Deep Research, OpenAI’s inference chip rumors, small VLMs and an AI agent job posting.

Paris AI Summit, Altman’s "Three Observations," Anthropic’s Economic Index

Live from Paris, Tim Hwang is at AI Action Summit 2025. In episode 42 of Mixture of Experts, we welcome Anastasia Stasenko, CEO and co-founder of pleais with our veteran experts. We analyze the Paris AI Summit, s1: Simple test-time scaling, Sam Altman’s “Three Observations,” and Anthropic’s Economic Index.

OpenAI's deep research and o3-mini, AI Action Summit and Anthropic’s Constitutional Classifiers

What does Sam Altman have up his sleeve? In episode 41 of Mixture of Experts, host Tim Hwang along with experts Nathalie Baracaldo, Marina Danilevsky and Chris Hay dissect OpenAI’s deep research and o3-mini, and the AI Action Summit. They also discuss Anthropic’s Constitutional Classifiers and Microsoft’s unit to study AI’s impact.

Watch all episodes from Mixture of Experts

Learn more about AI

What is artificial intelligence (AI)?

Applications and devices equipped with AI can see and identify objects. They can understand and respond to human language. They can learn from new information and experience. But what is AI?

What is fine-tuning?

It has become a fundamental deep learning technique, particularly in the training process of foundation models used for generative AI. But, what is fine-tuning and how does it work?

Build an AI-powered multimodal RAG system with Docling and Granite

In this tutorial, you will use IBM's Docling and open source IBM Granite vision, text-based embeddings and generative AI models to create a RAG system.

Stay on top of the AI news with our experts
Subscribe to our playlist on YouTube