Offline AI – Using Data Securely “On Premises”

This 1.5-hour course empowers participants to harness the power of AI while maintaining data security and privacy by focusing on offline, “on premises” deployments. We begin by defining offline AI and its advantages, particularly in sensitive data handling. The first half delves into LM Studio, a user-friendly tool for running large language models locally on standard computers, covering installation, model management, and practical demonstrations. The second half shifts to deploying AI on the Raspberry Pi, emphasising its portability and low-power consumption. We explore lightweight model alternatives, security hardening, and practical applications. Throughout, the course stresses data security best practices, ensuring participants understand how to process information locally without relying on cloud services.

Monday 7th April – 1.30pm – 3:00pm

Free

• Course Duration: 1.5 hours
• Focus: Harnessing AI while maintaining data security and privacy through offline, “on-premises” deployments.
• Introduction to Offline AI: Defining offline AI and its advantages for sensitive data handling.
• First Half of Course:
  • Introduction to LM Studio for running large language models locally.
  • Covers installation, model management, and practical demonstrations.
• Second Half of Course:
  • Focus on deploying AI on Raspberry Pi.
  • Emphasis on portability and low-power consumption.
  • Exploring lightweight model alternatives.
  • Security hardening techniques and practical applications.
• Data Security Focus: Stresses best practices for processing data locally without cloud reliance.