The Illusion of Agreement with ChatGPT: Sycophancy and Beyond
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908905818292224 |
|---|---|
| author | Noshin, Kazi Sultana, Sharifa |
| author_facet | Noshin, Kazi Sultana, Sharifa |
| contents | While concerns about ChatGPT-induced harms due to sycophancy and other behaviors, including gaslighting, have grown among researchers, how users themselves experience and mitigate these harms remain largely underexplored. We analyze Reddit discussions to investigate what concerns users report and how they address them. Our findings reveal five distinct user-reported concerns that manifest across multiple life domains, ranging from personal to societal: inducing delusion, digressing narratives, implicating users for models' limitations, inducing addiction, and providing unsupervised psychological support. We document three-tier user-driven suggestions spanning functional usage techniques, behavioral approaches, and private and institutional safeguards. Our findings show that AI-induced harms require coordinated interventions across users, developers, and policymakers. We discuss design implications and future directions to mitigate the harms and ensure user benefits. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_21409 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | The Illusion of Agreement with ChatGPT: Sycophancy and Beyond Noshin, Kazi Sultana, Sharifa Human-Computer Interaction While concerns about ChatGPT-induced harms due to sycophancy and other behaviors, including gaslighting, have grown among researchers, how users themselves experience and mitigate these harms remain largely underexplored. We analyze Reddit discussions to investigate what concerns users report and how they address them. Our findings reveal five distinct user-reported concerns that manifest across multiple life domains, ranging from personal to societal: inducing delusion, digressing narratives, implicating users for models' limitations, inducing addiction, and providing unsupervised psychological support. We document three-tier user-driven suggestions spanning functional usage techniques, behavioral approaches, and private and institutional safeguards. Our findings show that AI-induced harms require coordinated interventions across users, developers, and policymakers. We discuss design implications and future directions to mitigate the harms and ensure user benefits. |
| title | The Illusion of Agreement with ChatGPT: Sycophancy and Beyond |
| topic | Human-Computer Interaction |
| url | https://arxiv.org/abs/2603.21409 |