What happens when technology, intimacy and privacy intersect? Explore original member surveys, platform trends and data-led perspectives on modern relationships — with sources and limitations in plain sight.
A survey of 1,087 adult Sasha7 users in the US, Australia and Canada.
Start with two recent investigations into the increasingly blurred boundaries between digital tools, anonymity and intimacy.
reported using AI in connection with a discreet relationship
From writing messages to relationship advice, a voluntary survey of adult members asks where AI fits into discreet connections and private decision-making.
profile updates during the observed 2025 Halloween period versus October baseline
Changes in 2025 platform activity, paired with responses from 2,024 members surveyed ahead of Halloween 2026, offer a view of seasonal dating behavior.
Concise, attributable findings from Sasha7 studies. Copy a citation or go directly to the original report for definitions and methods.
Interpretation matters: These are results from Sasha7 members or Sasha7 platform activity, not representative estimates of infidelity among the general public. Survey answers are self-reported. Different studies use different samples, dates and denominators. How to interpret our research ↓
Browse 13 Sasha7-authored publications, including original surveys, platform analysis, community reporting and research-informed psychology explainers. These are different evidence types, not 13 independent scientific studies. Every link works without JavaScript.
Read the evidence label: Member surveys report self-reported responses; platform and regional reports describe Sasha7 activity; community overviews summarize public discussion; behavioral-science articles interpret previously published work. See interpretation notes ↓
How 1,087 adult members reported using AI for messaging, advice and handling private relationship circumstances.
Read source reportChanges in October 2025 platform activity alongside a 2026 survey of 2,024 members about costumes and anonymity.
Read source reportAn anonymous poll of 1,200 female members in the US and Australia about previous app use and privacy preferences.
Read source pollAn analysis of publicly visible activity and discussion patterns across selected online relationship communities.
Read source analysisA city ranking based on Sasha7's reported internal per-capita platform activity in the United States.
Read US reportA Portuguese-language ranking of Brazilian cities using reported platform activity relative to population.
Ler pesquisa em portuguêsA comparative ranking of seven Australian cities based on Sasha7's reported per-capita platform activity.
Read Australian reportAn examination of rising expectations within modern relationships, drawing on relationship psychology and the suffocation model.
Read analysisA discussion of attachment styles, communication and consensual non-monogamy. Interpretations are the author’s, not original clinical findings.
Read analysisExplores how consensual relationship agreements differ from infidelity, including themes of trust, jealousy and sexual dynamics.
Read analysisIntroduces research on novelty and sexual interest, while considering the limits of applying experimental findings to human infidelity.
Read analysisA thematic account of complaints discussed in selected affair-related subreddits. The published article does not provide an independently reproducible coding dataset.
Read analysisAn introductory synthesis of emotional, relational and situational factors associated with infidelity; not a new survey or causal experiment.
Read analysisAlso available in Brazilian Portuguese: AI and infidelity survey · Halloween research. These are language versions of existing studies, not additional independent datasets.
Figures are presented together with the population and comparison window they describe. Download the visible reference table as CSV — no account, tracking script or external service required.
| Finding | Result | Population / period | Source |
|---|---|---|---|
| AI use in connection with discreet relationships | 44% | 1,087 surveyed adult members, US/AU/CA, Jul–Sep 2026 | AI survey ↗ |
| Used AI to write or improve romantic messages | 31% | Share of all 1,087 survey respondents | AI survey ↗ |
| Used AI to explain suspicious circumstances | 18% | Share of all 1,087 survey respondents | AI survey ↗ |
| Used AI to make relationship harder to discover | 14% | Share of all 1,087 survey respondents | AI survey ↗ |
| AI made managing discreet relationship easier | 58% | Only respondents who had used AI, not all respondents | AI survey ↗ |
| Halloween-period change in profile updates | +81% | Oct 27–Nov 2, 2025 vs Oct 6–26, 2025 baseline | Halloween ↗ |
| Halloween-period change in messages | +47% | Equivalent Fri–Sun periods in 2025 analysis | Halloween ↗ |
| Halloween-period change in registrations | +28% | Halloween week vs comparison October week, 2025 | Halloween ↗ |
| Costumes help express personality while discreet | 62% | 2,024 adult members surveyed Oct 1–12, 2026 | Halloween ↗ |
| Female members who previously used Tinder | 72% | 1,200 surveyed women ages 31–49, US/AU, Jan–May 2026 | Women’s poll ↗ |
| Female members naming anonymity as top factor | 82% | Same poll of 1,200 female members, US/AU | Women’s poll ↗ |
We publish these findings to contribute to a more informed conversation about dating, intimacy and technology. Here's what the evidence does — and doesn't — establish.
Ask about a methodologyMember surveys reflect participating adults on Sasha7, a platform designed for discreet relationships. Participation may be voluntary and responses self-reported. These samples are not probability samples of all adults, married people, or people having affairs.
Registrations, messages and profile updates measure events on Sasha7. A percentage change compares platform activity in designated windows. It does not measure the prevalence of confirmed infidelity in a country or establish the cause of the change.
Not automatically. The AI survey (n=1,087), Halloween survey (n=2,024) and female-member poll (n=1,200) used different samples, eligibility criteria and field dates. Values reported among AI users, for instance, have a different denominator from values reported among all respondents.
US, Brazilian and Australian city rankings describe activity observed within Sasha7's own user base. They should not be presented as authoritative rankings of how often residents of those cities cheat. Check each original analysis for available sampling details and limitations.
The material in this library consists of Sasha7-produced member research, platform analyses and commentary. It should not be described as independently peer-reviewed academic research. Papers uploaded to Academia.edu are author-hosted, not evidence of independent peer review.
Each table entry links to the source article, which provides further context and, where available, field dates or methodologies. For questionnaires, underlying tabulations, permission to reproduce figures or additional detail, contact the research/press team. Only publicly reported summary results appear here.
Our behavioral-science features discuss external studies and offer editorial interpretations. They are not newly collected Sasha7 datasets or independent clinical trials. Where an article uses a coined framework or makes a hypothesis, treat it as interpretation rather than an established scientific finding. Community overviews likewise require clear sampling and coding methods before their conclusions can be treated as reproducible quantitative research.
Sasha7 Research is the public home for original surveys of Sasha7 members, reports on Sasha7 platform activity, selected public online-community analyses and data-led articles about dating, privacy and technology. Sasha7 operates the underlying adult dating platform and is the publisher of this research.
No. The 44% figure refers to respondents in a voluntary survey of 1,087 adult Sasha7 users based in the US, Australia and Canada. It is not a general-population estimate. Read the original AI study for the survey's definitions and methods.
No. It describes an association with changes in platform activity during a specified period in 2025, alongside a separate survey fielded in October 2026. It does not establish causation or track confirmed affairs. The full report specifies the comparison windows.
No. Behavioral-science articles interpret external research and may contain the author’s hypotheses. The original member surveys and platform analyses are separately labeled in the library. Cite the underlying scientific publications for academic claims, and cite Sasha7 for its own data or clearly attributed editorial analysis.
Yes. Cite the specific report and describe the sample accurately. This page includes copyable citation statements and a downloadable CSV of published summary figures. For full datasets, rights to reproduce graphics or methods beyond what is published, email [email protected].
Need a breakdown, sample definition, methodology clarification, visual assets, or comment? Get in touch with the Sasha7 press and research contact.
