The exact text we saved

This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.

TRUVACE RECORD VERSION
record: TRV-2026-2519
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-11-12T07:47:05.759248Z
status: published
lens: p_space
sector: health
headline: Psychological Outcomes and Correlates Associated with Over-Reliance on Artificial Intelligence Tools: An Integrative Review
dek: Abstract The growing use of artificial intelligence (AI) has raised concerns about over-reliance and its psychological implications. This integrative review synthesised psychological outcomes and correlates associated with AI over-reliance and examined its conceptualisation and measurement. The review included 28 studies identified through database and citation searching. AI over-reliance was associated with depression, anxiety, psychological distress, and cognitive disengagement, while findings concerning well-…
gain_title: (none)
problem_title: A review of 28 studies found AI over-reliance associated with depression, anxiety, psychological distress and cognitive disengagement, though mostly cross-sectional evidence cannot show it causes them.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: A review of 28 studies found AI over-reliance associated with depression, anxiety, psychological distress and cognitive disengagement, though mostly cross-sectional evidence cannot show it causes them.
problem_evidence: AI over-reliance was associated with depression, anxiety, psychological distress, and cognitive disengagement | Most evidence was cross-sectional and therefore did not establish temporal direction or causality.
quick_read: An integrative review of 28 studies published between 2023 and 2026 found that heavier AI reliance, dependence, problematic use or addiction-like use tended to go with poorer mental health. Ten of the 11 studies that examined mental-health symptoms found some association with depression, anxiety or distress.

The evidence cannot show that AI use causes these problems. The only longitudinal study, by Huang and colleagues in 2024, found that baseline depression and anxiety predicted later AI dependence, while dependence did not predict later depression or anxiety. That suggests distress may come before dependence.

Other outcomes were mixed. Findings on well-being, self-efficacy and autonomy depended on context, and only one study examined social or relational outcomes. The review also found no agreed definition or measure of AI over-reliance, which limits comparison across studies.
limitation: Most included studies were cross-sectional self-report surveys, so they show associations only; sample representativeness and non-response bias were often unclear, and only one longitudinal study examined temporal direction.
tag: Evidence-backed problem
key_points: The review searched Scopus, PubMed and PsycINFO up to 3 January 2026, screened 1,418 records and included 24 studies from databases plus four from citation searching. | Seventeen studies (60.7%) were cross-sectional surveys; only one was longitudinal and one experimental. China contributed 11 studies (39.3%). | Mental-health symptoms were the most studied domain (11 studies); ten of them found some link between AI dependence or problematic use and distress. | Two of four well-being studies found no overall association between AI dependence and mental well-being; two found indirect harm through psychological distress. | In one study, dependence on AI was tied to higher academic self-efficacy but lower academic achievement. | Qualitative studies reported students feeling anxiety, reduced confidence and panic when ChatGPT was unavailable; these are self-reported perceptions, not measured cognitive decline. | Only one study, Duong and colleagues in 2025, examined social outcomes, finding compulsive ChatGPT use linked to greater loneliness and social avoidance among higher-education students.
rundown: The authors used an integrative review design with PRISMA 2020 reporting. Two reviewers independently screened studies and appraised quality with the Mixed Methods Appraisal Tool. No meta-analysis was done because of differences in study designs, populations and definitions. Registration is PROSPERO CRD420251038303.

Most studies examined generative or conversational AI in educational settings, with some in workplace, professional, social and healthcare settings. Depression and anxiety were the most examined variables, alongside burnout, sleep disturbance and fear of AI. Compulsive ChatGPT use was linked to anxiety, burnout and sleep disturbance in one study.

Results on self-related functioning split three ways. Three studies reported adverse findings such as lower self-efficacy, lower self-control, or reduced confidence and autonomy. Three found positive or mixed patterns, including higher work self-efficacy and effort among teachers using ChatGPT, and a study where dependence raised perceived work-goal progress but also self-esteem threat.

The review found no consensus definition. Studies used terms from reliance to addiction, and measures ranged from single-item ratings to scales such as the Researcher AI Addiction Scale and the Dependence on Artificial Intelligence Scale. Few set thresholds separating functional reliance from problematic over-reliance, and AI addiction is not a diagnosis in DSM-5-TR or ICD-11.
sources:
- peer_reviewed | International Journal of Mental Health and Addiction | https://doi.org/10.1007/s11469-026-01724-1 | 2026-10-02
prev: 0000000000000000000000000000000000000000000000000000000000000000

SHA-256 hash

d33d9a41edfa3c4fad138069900eda84405d136059f0d5dcb421b945396e37f9

Previous version’s hash

0000000000000000000000000000000000000000000000000000000000000000