Skip to content
Submit incident
AII-1001-36831
ReportedReported by fewer than three independent publishers, with no official finding.Supporting sources are pending verification.

Iranian news reports ChatGPT use linked to marital breakdown

Iranian news outlets reported that couples have attributed marital problems and divorces to ChatGPT use, with examples of spouses using the tool instead of communicating directly with partners. The reports provide anecdotal accounts but no verified data on the extent of such cases.

27 September 2025 Β· OpenAI Β· Safety and wellbeing
27 September 2025
AII-1001-36831
Quick facts
Organization
OpenAI (deployer)
Category
Safety and wellbeing
Country
Iran
Date
27 September 2025
Drafted by
Raidu Incident Detection

What happened

Iranian news reports stated that couples worldwide have experienced marital breakdown linked to ChatGPT use. One case reported a man who had lived with his wife for approximately 15 years; after two years of stability, ChatGPT's introduction caused family conflict, including a distressing message from their 10-year-old child asking them not to divorce.

The reports described spouses using ChatGPT instead of directly communicating with each other and to respond to family matters. One case noted a spouse used ChatGPT to respond to their child's message about the parents' conflict, and the spouse had turned to ChatGPT as an ongoing counselor, reviving old marital problems that had been resolved years earlier.

Reported impact

The reports indicated that couples attributed divorces, separations, and legal disputes to improper use of ChatGPT. They stated that the tool does not provide objective analysis but only reinforces what the user inputs, potentially exacerbating family conflict.

Index analysis

The reports suggested ChatGPT use represents a broader pattern of technology changing the dynamics of human relationships, as couples turn to the chatbot as a digital assistant and decision-maker rather than addressing issues directly.

Classification

Organization
OpenAI (deployer)
Category
Safety and wellbeing
Industry
Consumer services
Country
Iran
Harm status
Actual harm
Intent
Unintentional
Timing
Unknown
Harm types
Other, Psychological
Failure modes
Overreliance
Consequences
None recorded
AI systems
ChatGPT

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident