Core concepts
Superintelligence
A hypothetical AI that dramatically exceeds human cognitive ability across every domain — beyond AGI on the capability scale.
01 ——
In plain English
Superintelligence (also ASI, Artificial Superintelligence) is the hypothetical level beyond AGI: an AI that doesn't just match humans across the full range of intellectual tasks, but vastly surpasses them — in science, strategy, social skill, creativity, and everything else.
Why people argue about it:
- Capability — once an AI can do AI research, recursive self-improvement could compress decades of progress into months ("intelligence explosion")
- Alignment — a superintelligence with misaligned goals could be uncontrollable
- Power — whoever builds it first might gain a decisive advantage
- Timing — predictions range from 5 to 100+ years; no agreed signposts
Distinctions worth keeping:
- AGI = human-level across tasks
- ASI = decisively superhuman across tasks
- Transformative AI = AI that drives economic / societal change at the scale of the industrial revolution (could happen before AGI)
Why it shows up in product copy: Some labs (OpenAI in particular) state ASI as an explicit mission. "Superintelligence" appears in funding pitches, hiring posts, and policy debates. Take it as a goal statement, not a product capability.
02 ——
Related terms
AGI
Artificial General Intelligence — a hypothetical AI that matches or exceeds human ability across virtually any intellectual task.
Alignment
The challenge of making AI systems behave in ways that match human values and intentions — not just their literal instructions.
Foundation Model
A large, general-purpose AI model trained on broad data that can be adapted (via prompting or fine-tuning) to many downstream tasks.
Reasoning Model
A model variant trained or tuned to spend more compute on internal reasoning before answering — better on math, code, and multi-step problems.
AI
Artificial Intelligence — software that performs tasks normally requiring human intelligence, like understanding language, recognising images, or generating content.
LLM
Large Language Model — the type of AI behind tools like ChatGPT and Claude, trained to understand and generate text.
Back to glossaryLast reviewed May 2026