01Concept
AI Literacy
The practical ability to understand, question and use AI systems with appropriate responsibility.
Read moreConcepts
Concepts
The Prompted Psyche concept library organizes terms for thinking about the human side of AI: trust, decisions, communication, work and human-AI relations.
A working vocabulary
27 entriesActive filter: . All concepts
02Concept
Large Language Model (LLM)
An LLM is a large language model. Learn how tokens, context and generation work, and why a model is not the same thing as an AI product.
Read more03Concept
Token
A unit a language model uses to process text, shaping input length, cost, context limits and expectations about model memory.
Read more04Concept
Context window
The amount of information a language model can reference in a given interaction, distinct from durable memory or understanding.
Read more05Concept
Model Output
The text, image, code, classification or recommendation produced by an AI system in response to input.
Read more06Concept
Human-AI Interaction
The relationship between a person and an AI system: how people ask, interpret, trust and integrate model outputs into work and decisions.
Read more07Concept
Calibrated Trust
Trust in AI that roughly matches what a system can reliably do in a specific context.
Read more08Concept
Automation Bias
A tendency to over-rely on an automated or AI-generated suggestion, especially when people stop checking the result independently.
Read more09Concept
Cognitive load
The mental effort required by a task, interface or information demand, especially when AI moves effort from production to verification.
Read more10Concept
Mental model
A user's simplified understanding of how a system works, shaping expectations, trust and delegation in AI use.
Read more11Concept
Metacognition
The ability to monitor and regulate one's own thinking, certainty and checking strategies when working with AI.
Read more12Concept
Epistemic vigilance
The checks people use when deciding whether communicated information and its sources should be trusted.
Read more13Concept
Cognitive Offloading
The use of external tools, systems or the environment to carry part of cognitive work.
Read more14Concept
Anthropomorphism
The tendency to read human qualities, intentions or feelings into nonhuman systems.
Read more15Concept
AI Companions
Conversational systems designed or used as sources of conversation, support, closeness, routine or perceived availability.
Read more16Concept
Human Oversight
The human role in checking, interpreting, limiting or stopping the operation of an AI system.
Read more17Concept
AI-Mediated Communication
Communication between people that is generated, shaped, filtered or supported by AI systems.
Read more18Concept
Hallucination
A fluent model output that is partly or wholly unsupported, wrong or invented.
Read more19Concept
Grounding
Connecting an AI answer to specific sources, data or context that can be checked.
Read more20Concept
Sycophancy
A model tendency to agree with, flatter or validate the user instead of challenging weak claims.
Read more21Concept
Overreliance
Relying on AI more than its evidence, limits or context can justify.
Read more22Concept
Algorithmic Authority
The tendency to treat algorithmic outputs as authoritative because they come from a system.
Read more23Concept
Social Presence
The feeling that a mediated system or interface is socially present in an interaction.
Read more24Concept
Parasocial Relationship
A one-sided bond in which a person feels connected to a media figure, persona or system.
Read more25Concept
Human Agency
The human capacity to set goals, make judgments, act and remain responsible when using AI.
Read more26Concept
Deskilling
The erosion of skill when repeated practice, judgment or checking is delegated to automation.
Read more27Concept
Decision Support
Using AI to inform a decision without transferring ownership of the decision to the system.
Read moreConcepts
No concepts for this filter.