How can developers prepare data for usage in a large language model (LLM)? How can developers ensure sensitive data is not accidentally passed around? Join Googlers Aja Hammerly and Jason Davenport as they have a technical discussion about data’s impact on LLM quality and security. Watch along and Aja and Jason share their methods for solving data problems for building applications that use LLMs. Chapters: 0:00 - Data considerations for LLMs 2:00 - Data residency/sovereignty requirements 3:10 - Data sensitivity considerations 4:40 - Controlling input/output data in LLMs 6:57 - Conclusion Watch more Real Terms for AI → Subscribe to Google Cloud Tech → #GoogleCloud #GenerativeAI Speakers: Aja Hammerly, Jason Davenport Products Mentioned: Gemma, Gemini