Ben Lorica and Paco Nathan on New LLMs, Insights from GTC 2024, Constraint-Driven Innovation, Model Safety, and GraphRAG.
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: recently released large language models, Constraint-Driven Innovation, highlights from GTC 2024, and Lessons from the First AI Workload Security Exploit.
Relevant Links:
- LLM models: Databricks DBRX ; AI21Labs Jamba ; SambaNova Samba-CoE ; The Efficient Frontier of LLMs
- Constraint-Driven Innovation: This presentation argues that the removal of traditional constraints in database systems, driven by cloud computing, open source, and hardware specialization, has ushered in a golden age of database innovation, with machine learning playing a central role and hardware specialization offering further opportunities.
- Nvidia’s GTC 2024 Announcements: a cheat sheet
- First Known Attack Campaign Targeting AI Workloads Actively Exploited In The Wild: Highlights a critical security risk for AI teams, where vulnerabilities in AI infrastructure, like the Ray framework, can expose sensitive data and computing resources to attackers. This underscores the need for AI teams to adopt proactive security measures, including runtime monitoring, best practices, and mitigation strategies. The incident reflects the growing importance of AI security as the technology becomes more widely adopted and targeted by malicious actors.
- Is Your Data Strategy Ready for Generative AI?
- AI safety is not a model property
- The Intersection of LLMs, Knowledge Graphs, and Query Generation
- Recommendation: Death of Effective Altruism
Interview highlights – key sections from the video version:
- Newly released LLMs and trends in large language models
- Constraint-Driven Innovation & what it means for LLMs
- Nvidia’s GTC 2024 Announcements
- First Known Attack Campaign Targeting AI Workloads
- Data tools for Generative AI
- Model safety
- GraphRAG
Related content:
- A video version of this conversation is available on our YouTube channel.
- Ben Lorica and Paco Nathan: Frontiers of AI – From Text-to-Video Models to Knowledge Graphs
- Semih Salihoglu: The Intersection of LLMs, Knowledge Graphs, and Query Generation
- 2024 Themes and Trends in AI
If you enjoyed this episode, please support our work by encouraging your friends and colleagues to subscribe to our newsletter:
