In the model, visible unverified peer use (social proof) suppresses verification and tips the population into collective overreliance, via a smooth crossover with no hysteresis¶

Claim · All claims
Evidence · 1 study (2 entries) · theoretical r3 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? As social proof strength s rises from 0 to 0.6, verification falls from 0.29 to 0.002 and overreliance rises from 0.30 to 0.52, with hub and random networks coinciding. → Ahana Biswas 2026 q2 i? Forward and backward sweeps coincide at every s even at θ = 30, so the ABM shows a smooth crossover with no hysteresis. → Ahana Biswas 2026 (2)

Evidence¶

Ahana Biswas 2026¶

Ahana Biswas. (2026). Modeling AI Overreliance as a Complex Adaptive System. arXiv. https://arxiv.org/abs/2608.19616 doi:10.48550/arxiv.2608.19616

q2 · i? · theoretical · r3

Simulation sweep of the social-proof strength s from 0 to 0.6 (Fig. 4), engaging the verification-suppression feedback in which visible unverified peer use raises the utility of unverified AI use. "verification falls 0.29→0.002 (<0.01 =\"collapse\") and overreliance rises 0.30→0.52".

"assrises 0→0.6, verification falls 0.29→0.002 (<0.01 ="collapse") and overreliance rises 0.30→0.52 (Fig. 4); hub and random coincide within CIs."

Ahana Biswas 2026 (2)¶

Ahana Biswas. (2026). Modeling AI Overreliance as a Complex Adaptive System. arXiv. https://arxiv.org/abs/2608.19616

q2 · i? · theoretical · r3

Hysteresis test of the cascade: forward and backward sweeps from low- versus high-trust starts across social-proof strengths, including θ = 30. The ABM shows "a smooth crossover withno hysteresis"; the mean-field saddle-node of Proposition 3 appears only under strong coupling.

"Forward/backward sweeps (low- vs high-trust start)coincide at everys even atθ= 30: the ABM shows a smooth crossover withno hysteresis; the saddle-node (Prop. 3; mean-field fold in the supplement) is the strong-coupling analogue."

Discussion¶

Related Claims¶

• In the model, social learning creates consensus in trust without shifting aggregate reliance; connectivity alone does not produce collective overreliance — related
• In the model, interventions that alter the feedback channel (verification visibility, social-proof damping) reduce overreliance and regret, while lowering verification cost alone does not lower regret — related
• In the agent-based model, task difficulty and AI quality set the baseline level of overreliance, with difficulty the dominant global factor — related
• Classroom verification norm is predicted to show threshold behavior and hysteresis rather than a smooth response to teacher-set requirements — reports the opposite
• In the model, topology shifts aggregate reliance only under opinion dynamics, when influential hubs transmit correlated beliefs that move agents across the verify/use margin — related