Augmented Intelligence

Augmented Intelligence is the framing that AI should extend human capability rather than be treated as an autonomous truth source. In EP 47: The AI Pioneer Who Decided Privacy Matters More Than Hype, Jonathan Schaeffer prefers this phrase because LLMs can be useful like graduate students or interns while still requiring supervision, verification, and responsibility from the human user.

The concept is a concise bridge between Human Judgment Under AI, AI Verification, and AI Hallucination. It accepts that AI can accelerate reading, drafting, retrieval, and analysis, but it treats unchecked delegation as the failure mode.

Key Claims

• Augmented intelligence keeps the user accountable for deciding what problem matters, what evidence is enough, and whether output can be used.
• The frame fits probabilistic LLMs better than language that implies the system knows, intends, or understands in a human way.
• It pairs naturally with Local Private AI when the goal is to make private data more useful without surrendering human review.

Connections

• Jonathan Schaeffer and Data Science With Sam - source context.
• Human Judgment Under AI, AI Verification, and AI Hallucination - reliability and responsibility context.
• Kind Private AI, Retrieval-Augmented Generation, and AI Professional Data Security - supervised private-AI use cases.