We explore the Rejection Spiral, a psychological phenomenon where users swing from over-reliance on AI to total abandonment after encountering a significant error. Because conversational AI uses social cues and fluent language, its mistakes often feel like a personal betrayal rather than a simple technical bug.
This emotional backlash leads to global discounting, where people ban the technology entirely, inadvertently removing the safety rails and productivity benefits the tool once provided. We argue that blanket bans are counterproductive because they drive AI usage underground and prevent users from developing calibrated trust. To break this cycle, we suggest moving toward repairable relationships where failure is treated as a prompt for better verification rather than a reason for complete withdrawal.
Overall, we advocate for a balanced partnership with AI that acknowledges its unreliability while maintaining human agency.
Full essay available here
