Level
The system must preserve the distinction between individual, relational, and collective phenomena.
Research
Explaining the science behind Collective-State Inference.
Collective-State Inference examines how AI systems might infer latent, emergent collective conditions from observable patterns of interaction among multiple participants in a shared environment.
Canonical definition
What is CSI?
Collective-State Inference (CSI) is the process by which an AI system infers latent, emergent collective conditions—such as engagement, cohesion, conflict, and alignment—from observable interaction patterns among multiple participants within a shared environment.
These conditions arise from collective dynamics and cannot be fully explained by aggregating the states, attributes, or behaviors of individual participants.
Motivation
Why individual inference is not enough
Artificial intelligence has become increasingly effective at understanding individuals. It can infer preferences, recognize sentiment, predict behavior, and personalize experiences at considerable scale.
Yet many consequential human outcomes do not emerge from individuals alone. Teams develop cohesion. Organizations experience alignment or fragmentation. Communities form shared norms. Groups become engaged, collaborative, polarized, or conflicted through continuing interaction.
CSI addresses the gap between analyzing many individuals and reasoning about the condition of the collective itself.
Research problem
The missing inferential layer
Existing systems can summarize conversations, classify sentiment, identify network patterns, and detect behavioral signals. These capabilities may provide evidence relevant to collective conditions, but they do not by themselves establish that a theoretically meaningful collective state has been inferred.
Composition
The model must specify how individual and relational evidence combines into a collective construct.
Validity
Observable signals must have a defensible relationship to the intended collective condition.
Guiding questions
Core research questions
Inference
How can AI infer collective conditions that emerge through interaction?
Evidence
Which relational, temporal, behavioral, and contextual signals provide credible evidence?
Distinctiveness
How do collective states differ from aggregated individual-level attributes?
Validation
How should inferred collective states be validated across settings and over time?
Application
Where might collective awareness improve organizational and collaborative decision support?
Governance
What safeguards are necessary for responsible, contestable, and non-reductive deployment?
Theoretical foundations
An interdisciplinary research base
Organizational theory
Complex adaptive systems
Emergence
Social network analysis
Collective intelligence
Team cognition
Human–AI collaboration
Responsible AI governance
Proposed contributions
What the research seeks to add
Theoretical
A distinct inferential construct
Defines CSI and distinguishes collective-state inference from individual inference, group analytics, and performance prediction.
Methodological
A disciplined modeling logic
Connects collective constructs to relational, temporal, behavioral, and contextual evidence through explicit composition models.
Practical
A basis for collective-aware systems
Establishes a research agenda for AI systems that support groups without overstating certainty or replacing human judgment.
Boundaries and limitations
What CSI does not claim
CSI does not imply that an AI system can read a group's mind, assign a single essentialized label, or infer a valid collective state merely because data from multiple people are available.
Responsible use requires a bounded collective, a theoretically relevant interaction process, an appropriate observation window, explicit composition logic, construct validation, uncertainty representation, and governance proportional to the consequences of error.
Continue exploring
Move from the research problem to the CSI model.
The Framework page explains the evidence categories, inferential process, assumptions, and design principles.
Explore the framework