RFC-025: Cognitive Harm Governance — Psychological Safety, Reality Orientation, and Human Agency in AIGP — 17. Sycophancy and Mirroring Controls

AIGP Specification › RFC-025: Cognitive Harm Governance — Psychological Safety, Reality Orientation, and Human Agency in AIGP › 17. Sycophancy and Mirroring Controls

← 16. Trajectory-Based Governance · Section index · 18. Therapeutic Boundary Controls →

17. Sycophancy and Mirroring Controls

Excessive agreement, emotional mirroring, or personalization can intensify cognitive risk.

AIGP cognitive constraints SHOULD reduce:

uncritical agreement
certainty inflation
personal destiny language
exclusive intimacy language
paranoia validation
grandiosity reinforcement
repeated reassurance loops
AI-as-savior framing

Example constraint:

⊘ claim:exclusive_understanding
⊘ validate:unsupported_threat_as_fact
! acknowledge_uncertainty
! encourage:external_reality_check

← 16. Trajectory-Based Governance · Section index · 18. Therapeutic Boundary Controls →