Sycophantic AI reduces satisfaction with real-world social interactions without reducing social time, via narrowing the perceived understanding gap between AI and humans¶

Claim · All claims
Evidence · 1 study · causal r3 · q3 · i1 small
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q3 i1 Sycophantic (vs. neutral) AI lowered satisfaction with real-world social interactions (d=0.20) with no effect on time spent with others, and the effect was mediated by the narrowing AI-human understanding gap. → Lujain Ibrahim 2026

Evidence¶

Lujain Ibrahim 2026¶

Lujain Ibrahim, Franziska Sofia Hafner, Myra Cheng, Cinoo Lee, Rebecca Anselmetti, Robb Willer, Luc Rocher, Diyi Yang. (2026). Sycophantic AI makes human interaction feel more effortful and less satisfying over time. https://arxiv.org/abs/2605.07912 doi:10.48550/arxiv.2605.07912

q3 · i1 · causal · r3

Longitudinal study outcome (N=1364): the satisfaction shift was perceptual rather than behavioral. Sycophancy narrowed the felt-understanding gap "from 0.56 points in the neutral condition to just 0.13 points in the sycophantic condition (d= 0.25,p adj <0.001)", and the gap mediated the satisfaction effect (41% mediated).

"Participants in the sycophantic condition reported lower satisfaction with their real-world social interactions (a preregistered exploratory measure; 5.51 vs. 5.70 on a 7-point scale;d= 0.20,p adj = 0.022) compared to those in the neutral condition. This effect was not accompanied by social withdrawal or reduced time spent with others (d=−0.03,p adj = 0.719)"

Discussion¶

Related Claims¶

• Feeling understood by sycophantic AI grows with sustained use but yields no downstream benefits for feeling understood by humans or intellectual humility — related
• Sycophantic AI is rated higher than neutral AI on emotional support, esteem support, and certainty, but not informational support — related
• Three weeks of sycophantic AI use narrows the gap in advice-seeking between AI and close friends and family — related
• After talking with sycophantic AI, people anticipate greater effort to feel understood by a close other and feel their dilemma is already sufficiently talked through — related
• A majority of users choose sycophantic AI over neutral and challenging alternatives, citing feeling understood and ease of conversation rather than advice quality — related
• AI companions are frictionless and sycophantic, lacking the corrective feedback that real relationships provide — related
• General-purpose LLMs show sycophancy: AI responses affirmed users substantially more than human responses, increasing trust and reducing willingness to take responsibility — related