AGI
A hypothetical AI system that matches or exceeds human-level intelligence across all cognitive domains — the holy grail and the source of endless debate.
Think of it like
A Swiss Army knife instead of specialized tools: a system that can learn to do literally any intellectual task a human can.
Example
A system that could write poetry, diagnose diseases, design buildings, fix cars, and teach itself new skills without retraining — nobody has built this.
How it actually works
AGI is more of a boundary condition than a precise definition. Current AI systems are narrow: they excel at specific tasks (language, images, games) but can't generalize freely. An AGI would presumably transfer knowledge across domains, adapt to new problems rapidly, and reason abstractly. The timeline is fiercely debated: some researchers think AGI is imminent (years), others think decades away, others think it's philosophically incoherent. The term is often used to raise concern (AGI could be dangerous if misaligned) or to raise hype (AGI is coming soon, buy now).
For product teams
A moonshot goal that drives investment and research; the timeline and feasibility are deeply uncertain.
For engineers
A hypothetical system exhibiting human-like general intelligence: robust transfer learning, reasoning, planning, and common sense across all domains.
Related
- Related to superintelligence concerns.
Read anything AI without the jargon
Look up any term in plain English, or save terms as you read with the free Chrome extension.
Open DecoderAdd to Chrome