aceModerate evidence
More
Review of 188 records links compulsive AI chatbot use to distress and lower wellbeing
Study title: Problematic engagement with generative AI measurement drivers contexts and outcomes
- Evidence type
- Systematic review or meta-analysis
- Sample
- n = 188
- Sources
- 1 sourcePeer-reviewed
- Published
- 2 min read
- Truvace record
- TRV-2026-2507cite or check it
Study title: Problematic engagement with generative AI measurement drivers contexts and outcomes
A PRISMA-style review of research on problematic generative AI use started from 188 records found in a Scopus search. It found that compulsive or dependent use is associated with technostress, psychological distress and lower wellbeing. The authors treat addiction-like language as part of a broader problematic-use idea, not a clinical diagnosis.
Read the study abstract
Generative AI tools and conversational chatbots are increasingly embedded in education, work, and daily life, raising concerns about engagement that becomes difficult to self-regulate. This review synthesizes emerging research on problematic generative AI use, treating addiction-like language as part of a broader problematic-use construct rather than a clinical diagnosis. Following PRISMA 2020 reporting, a Scopus search executed identified 188 records. Included studies were characterised by research design and e…
The problem
A review of 188 records found that compulsive or dependent generative AI use is associated with technostress, psychological distress and lower wellbeing, though the studies are almost all cross-sectional and self-reported.
In brief
Studies point to several drivers: attachment, anthropomorphism and perceived empathy, loneliness and social anxiety, immersive flow, and strain on self-regulation. The review suggests social presence and immersion push some users toward heavier reliance. In learning and knowledge work, performance pressure and incentives can normalize overreliance, with reported risks to independent thinking, academic integrity and skill development.
The findings are associations, not proven causes, because the research is almost entirely cross-sectional and self-reported. The authors say the key line is between bounded, purpose-driven use and escalating reliance with reduced self-regulation, not between users and non-users.
Main points
- The review followed PRISMA 2020 reporting and began with 188 records from a Scopus search.
- Included studies were sorted by research design and evidence type, not scored with a formal risk-of-bias tool.
- Measurement instruments for problematic use remain partly fragmented, blurring high engagement, habitual reliance and impairment.
- Mechanism studies converge on attachment, anthropomorphism, perceived empathy, loneliness, social anxiety, immersive flow and self-regulatory strain.
- In learning and work, performance pressure and incentives can normalize overreliance, with reported risks for cognitive offloading and academic integrity.
- Mitigation discussion covers reinforcement-friendly interface cues, privacy and memory governance, and sustainability trade-offs.
- A sensitivity analysis comparing 2026 studies with earlier ones reversed no conclusion; recent longitudinal studies strengthened evidence on temporal order and outcomes.
The full story
The review maps a fast-growing field in which generative AI tools and chatbots are embedded in education, work and daily life. It frames the issue as problematic use rather than clinical addiction and notes that measurement tools are still fragmented.
On causes, the studies converge on emotional and social factors: attachment, seeing the chatbot as human-like, perceived empathy, loneliness, social anxiety and immersive flow. The authors suggest social presence and immersion amplify reliance for some users.
On consequences, compulsive or dependent use is associated with technostress, psychological distress and lower wellbeing. In education and knowledge work, reported risks include cognitive offloading, weaker independent thinking, academic integrity problems and skill development.
The authors use an Antecedents-Decisions-Outcomes structure only to frame future longitudinal and intervention research. A comparison of 2026 studies with earlier ones left the design profile and main gaps unchanged.
Sources
1 on file- Social Sciences & Humanities Opendoi:10.1016/j.ssaho.2026.103890
Talk it over
Who carries this cost, and what would have to change to prevent it?
Review of 188 records links compulsive AI chatbot use to distress and lower wellbeinghttps://truvace.com/item/problematic-engagement-with-generative-ai-measurement-drivers-contexts-and-outco
Who carries this cost, and what would have to change to prevent it?
The problem A review of 188 records found that compulsive or dependent generative AI use is associated with technostress, psychological distress and lower wellbeing, though the studies are almost all cross-sectional and self-reported.
Answer in the threadHow do we know? What can this kind of evidence show, and what can’t it?
The evidence This rests on a systematic review or meta-analysis with a sample of 188, from Social Sciences & Humanities Open.
Answer in the threadWhat does it leave open? What else would you want to know before acting on it?
The record’s own caveat The studies reviewed are almost entirely cross-sectional and self-reported, so the relationships are associations rather than causal effects.
Answer in the threadHave you seen AI used this way where you study, work or live? Did it go the same way?
Answer in the thread
The questions are the same on every record. The details under them are taken from this record as written.
What readers say