Episode 9: Guardian AI, hybrid AI agents, and brand-safe automation, with Brian Ritchie, CEO and Founder, kama.ai
In this episode of Responsible AI in Action, Brian Ritchie explains the emerging concept of Guardian AI and how it relates to kama.ai’s Sober Second Mind approach. He outlines the difference between today’s hybrid AI agents and more autonomous agentic AI, and why the industry’s move toward AGI and superintelligence raises new questions about control, safety, and human values.
From guardian “angel” oversight to the practical realities of deploying hybrid AI agents inside enterprises, Brian walks through how a knowledge graph guided by human values can act as a human behavioural model and governor for other AIs. He also describes Emotion Graph RAG, a patent-pending capability that can help evaluate both user inquiries and AI responses through a human-value lens before they reach the end user.
Watch the full episode now:
Key themes include:
-
What Guardian AI is and how it relates to Sober Second Mind AI
-
The rise of agentic AI and its proximity to artificial general intelligence
-
How kama.ai’s hybrid AI agent combines a deterministic “left brain” with a generative “right brain”
-
Emotion Graph RAG as a way to apply human values to AI inquiries and responses
-
How Guardian AI concepts can prevent harmful or high-risk outputs before they reach users
-
Turning responsible AI, governance, and human-value modeling into a brand and trust differentiator
-
Using “risk sliders” and audience controls to decide where generative AI can safely be exposed
Whether you are designing AI agents, responsible for governance and compliance, or looking for ways to make AI a trusted extension of your brand, this episode offers a clear and practical view of how Guardian AI can support safer enterprise adoption.
🎧 Listen now to explore how hybrid AI and Guardian AI can enable brand-safe automation at scale.
Learn more about Brian Ritchie and kama.ai
Website: kama.ai
Follow on LinkedIn: kama.ai | Brian Ritchie