From the course: Hands-On AI: Building Agents with the Google Agent Development Toolkit (ADK)
Join today to access over 25,000 courses taught by industry experts.
Responsible agents: Upholding safety, ethics, and trust
From the course: Hands-On AI: Building Agents with the Google Agent Development Toolkit (ADK)
Responsible agents: Upholding safety, ethics, and trust
- Imagine you've built WanderWise to help users plan amazing trips. But what if it starts sharing personal travel histories, suggests age inappropriate activities, or recommends unsafe areas? This isn't science fiction. It's why responsible AI agent development is paramount, especially with powerful tools. Building responsible agents that protect users, maintain trust and operate safely isn't about avoiding problems. It's about creating AI that genuinely enhances human experiences. First, agent authentication or agent-auth, the agent uses its own service account credentials. This is suitable when all users interacting with the agent require the same level of access to backend resources. Second, user authentication, or user-auth. The agent acts on behalf of an individual user, typically using OAuth 2.0 tokens obtained via a three-legged OAuth flow. This ensures the agent only accesses data the specific user is authorized to see. For WanderWise, integrating a calendar to add travel…
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.