---
id: "claim-ai-disrupts-coordination"
type: "claim"
source_timestamps: ["§ Where Coordination Breaks Down"]
source_url: "https://hbr.org/2026/02/how-to-foster-psychological-safety-when-ai-erodes-trust-on-your-team"
source_title: "How to Foster Psychological Safety When AI Erodes Trust on Your Team"
tags: ["coordination", "team-dynamics", "social-cues"]
related: ["concept-human-ai-oversight-paradox"]
speakers: ["Jayshree Seth", "Amy C. Edmondson"]
confidence: "high"
testable: true
sources: ["adoption"]
sourceVaultSlug: "hbr-seg-adoption"
originDay: 9
articleStem: "hbr-cl-79-psychological-safety-ai-trust"
sourceUrl: "https://hbr.org/2026/02/how-to-foster-psychological-safety-when-ai-erodes-trust-on-your-team"
sourceTitle: "How to Foster Psychological Safety When AI Erodes Trust on Your Team"
---
# AI Team Members Negatively Impact Human Coordination

**Claim (confidence: high · testable).** Recent studies indicate that introducing an AI "team member" can **disrupt vital group processes.**

The mechanisms the authors list: AI **fails to pick up on contextual cues**, **cannot adjust its communication style** based on team dynamics, and **does not engage in informal relationship-building.** Furthermore — a structural asymmetry — AI **suffers no social or professional consequences for being wrong.** This forces human teammates to work around an entity operating by *fundamentally different rules*, which **reduces human effort, impairs communication, and creates an accumulated cost to team coordination.**

This claim is the downstream effect of the [[concept-human-ai-oversight-paradox]] and motivates the fourth framework pillar, [[framework-ai-integration-principles|emphasizing human connection]] (preserving human-only space, [[action-create-override-protocols|override protocols]]).

**How to test it:** measure coordination cost, communication overhead, and role clarity in matched teams with vs. without an embedded AI "member."

**Enrichment (partially supported; context-dependent):** APA and Nature evidence align with the harm direction. **Important counter-evidence:** in *structured, well-defined* settings (manufacturing, logistics, software-engineering assistants) well-integrated AI agents can *improve* coordination, clarify roles, and centralize information. The synthesis: coordination harms are most acute in **ambiguous, social, high-context** work; in routinized settings with explicit protocols, AI can be a coordination *asset*.