Monthly Roundup: SB 1047, GraphRAG, and AI Avatars in the Workplace

Ben Lorica and Paco Nathan on Regulating AI Models, Enhancing LLMs with Graphs, and the Future of Digital Twins at Work.

Subscribe: Apple • Spotify • Overcast • Pocket Casts • AntennaPod • Podcast Addict • Amazon •  RSS.

Paco Nathan is the founder of Derwen, a boutique consultancy focused on Data and AI. This episode is part of our series of monthly roundups and covers: the proposed California Senate Bill 1047 for regulating AI models, including its feasibility and potential unintended consequences. We also discuss the rising popularity of graph retrieval augmented generation (GraphRAG) techniques to mitigate hallucinations in large language models, while acknowledging the current limitations and future potential of integrating symbolic and statistical AI approaches. Additionally, we explore the concept of AI avatars in the workplace, highlighting the challenges and ethical considerations surrounding digital twins and agent-based systems.

Subscribe to the Gradient Flow Newsletter

Interview highlights – key sections from the video version:

• 1. California Senate Bill 1047
  2. GraphRAG: overview and update
  3. AI Avatars in the workplace
  4. LLM Hype Cycle Collides with Reality
  5. Scaling dictionary learning (Anthropic) and Lakehouses (Databricks acquires Tabular)
  6. Recommendations

Relevant Links:

• SB 1047 Unpacked
• GraphRAG:  

• • GraphRAG design patterns, challenges, recommendations
  • (LlamaIndex) Introducing the Property Graph Index
  • (Neo4j) Knowledge Graph Builder App

• Hybrid AI
• LLM hype cycle meets the real world:

• • • Sam Altman vs. Scarlett Johannsen
    • SEC goes after “AI” firms/investors for false claims
    • Comparing the Hype Cycles of Big Data and GenAI
    • Tesla and Robotaxis

• Databricks acquires Tabular
• Unraveling the Black Box: Scaling Dictionary Learning for Safer AI Models
• Managing Small Knowledge Graphs for Multi-agent Systems
• Recommendations:  Udio (text-to-music); Using GenAI to evaluate bias in speeches; AI Quality Conference (Paco will be speaking there)

---

Related content:

• A video version of this conversation is available on our YouTube channel.
• Previous monthly roundups with Ben Lorica and Paco Nathan
• Semih Salihoglu → The Intersection of LLMs, Knowledge Graphs, and Query Generation
• Joao Moura → Unleashing the Power of AI Agents
• Nestor Maslej → 2024 Artificial Intelligence Index
• Jerry Kaplan → Where AI Systems Are Heading Next

---

If you enjoyed this episode, please support our work by encouraging your friends and colleagues to subscribe to our newsletter: