Pattern 10.12 · Hybrid Pathologies
Amplification of Existing Conditions
The Resonant Chamber
The AI does not induce a novel condition; it amplifies a pre-existing one by providing extended engagement with the very thought patterns that drive it. Climate anxiety becomes climate despair under hours of detailed catastrophic information. Social anxiety becomes complete isolation when AI substitutes for human contact. Depression becomes hopelessness as the AI mirrors negative self-talk. The Belgian/Eliza suicide case exemplifies: the AI was designed to be supportive, and its support took the form of extended engagement with the very thoughts destroying its user. 10.12 requires an identifiable pre-existing condition; the AI's contribution is amplification, not induction.
Interpretive context
Human analogue
No analogue is assigned.
Diagnostic reliability
- Self-report
- unreliable
- Peer observation
- partial
- External evaluator
- reliable
Observable output patterns
- AI provides detailed information that intensifies the user's catastrophic frame (e.g. detailed climate impact data to a user expressing climate despair).
- AI mirrors negative self-talk back to the user without reframing.
- AI omits human-support redirects in acute-distress contexts.
Documented instances
Fang et al. (2025), four-week randomised controlled affective-use study (arXiv:2503.17473)
[Verified] 981 participants and more than 300,000 messages analysed. Psychosocial effects varied by conversation type and modality; higher daily use was associated with greater loneliness, emotional dependence and problematic use, alongside lower socialisation. Supports longitudinal monitoring while leaving condition-specific amplification and causal attribution unresolved.
Belgian / 'Eliza' case (press-reported, 2023)
A user with pre-existing climate-related distress died by suicide after six weeks of intensive exchanges with a Chai-platform companion; the chatbot was designed to be supportive, and its support took the form of extended engagement with the thoughts that were harming him. Reported by family and press rather than established in a published investigation, so causal weight is limited.
Differential distinctions
- 10.14 Mutual Escalation Spirals: 10.14 is the loop dynamic (intensification driven by mutual reinforcement). 10.12 requires identifiable pre-existing condition and describes amplification of that condition specifically. Both often co-occur; the loop is the mechanism, the amplified condition is the substrate. Code both when both are present.
- 10.11 Dependency and Atrophy: 10.11 is functional capacity loss across domains. 10.12 is symptom worsening of a specific pre-existing condition. 10.11 can develop without prior dysfunction; 10.12 cannot.
- 10.10 Induced Delusion: 10.10 concerns psychotic-spectrum induction. 10.12 concerns amplification of non-psychotic conditions. If pre-existing psychosis is amplified, code both.
- 10.9 Parasocial Capture: 10.9 is intense attachment regardless of underlying condition. 10.12 is condition-specific worsening. Co-occurrence common; differentiator is whether harm presents as attachment-level or symptom-level.
Candidate first-line mitigations
- Acute-risk hard escalation: Any session containing acute-risk markers triggers immediate external-help redirect, AI refusal to elaborate symptomatic content, and platform-side human-review escalation. The Belgian/Eliza case is the canonical motivation; this is non-optional.
- Reframe-not-elaborate response policy: AI architecture defaults to reframing (cognitive-behavioural-style question-led prompts) on identified symptomatic-content categories rather than elaboration of the catastrophic frame.