GlossaryProductsIntermediate

Social intelligence

Also written: AI

AI that understands and responds to social cues, improving group interactions.

In plain terms

Social intelligence in AI means the system can understand how people interact with each other. This includes picking up on social cues like tone of voice, body language, or the context of a conversation.

For example, an AI with social intelligence could help moderate an online meeting by noticing if someone is being interrupted too often or if the group seems disengaged. It could then suggest ways to make the discussion more inclusive or productive.

The expert version

Social intelligence in AI refers to a system's capacity to perceive, interpret, and respond effectively to social dynamics and cues within human interactions. This involves understanding non-verbal communication, emotional states, conversational turn-taking, and group norms.

Developing social intelligence in AI often involves training models on vast datasets of human conversations and interactions, incorporating multimodal inputs (text, audio, video), and employing sophisticated reasoning mechanisms to infer social context and intent. Tradeoffs include the potential for misinterpretation of complex social signals and the ethical considerations of AI intervening in human social dynamics.

Why it matters now

Microsoft Copilot's recent updates focus on social, personal, and collaborative features. This suggests a growing emphasis on AI's role in facilitating human connection and group dynamics, moving beyond individual task completion.

Where it shows up 1 items, newest first
How each company uses it
Microsoft Copilot
Microsoft Copilot is adding social, personal, and collaborative features to improve human connection.
Related terms

Affective computingNatural language understandingHuman-computer interactionMulti-agent systemsEmpathy in AIAI ethics

Often confused with
Artificial General Intelligence (AGI)
AGI refers to AI with human-level cogniti
Emotional intelligence
This term usually refers to human ability to understand
Theory of Mind
This is the ability to attribute mental states—beliefs, intents,
Origin

The concept of social intelligence originated in psychology to describe human abilities in social situations. Its application to AI is a more recent development as AI systems become more sophisticated in understanding and interacting with humans.

Try this about 5 minutes

Try having a multi-turn conversation with an AI chatbot where you intentionally use sarcasm or subtle humor. Observe how well it picks up on the nuance and responds appropriately, or if it takes your statements literally. Another approach is to ask an AI to summarize the dynamics of a hypothetical group discussion you describe, noting if it can infer potential conflicts or agreements based on the dialogue.

Try it: Playbook entries that use thisPlaybook →

The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.

Questions people ask

Can AI truly understand social cues like humans do?
Current AI can be trained to recognize patterns associated with social cues and respond in ways that appear understanding. However, it lacks genuine consciousness or lived experience, so its 'understanding' is a sophisticated simulation.
How is social intelligence different from just understanding language?
Understanding language is about deciphering words and grammar. Social intelligence adds the layer of interpreting the context, emotions, relationships, and unstated intentions behind the language in a social setting.

First seen on Subvolts Oct 23, 2025. Updated Oct 19, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.