Cyberbullying Statistics 2026
By Axis Intelligence Research & Marcus Chen | Last updated: June 13, 2026 | Next scheduled update: Q3 2026 (September)
Quick Answer:
58% of U.S. teens aged 13–17 experienced cyberbullying in their lifetime in 2025 — the highest rate the Cyberbullying Research Center has recorded in nearly two decades of measurement. Globally, the WHO/Europe HBSC study puts the figure at 1 in 6 adolescents across 44 countries. AI-generated abuse, financial sextortion, and deepfake harassment have made 2025–2026 the most dangerous digital environment for minors in recorded history.
Key Findings
- 58% of U.S. teens have experienced cyberbullying in their lifetime (CRC 2025, n=3,466) — up from 33.6% in 2016, a 73% increase over nine years.
- 1 in 6 adolescents globally has been cyberbullied, per the WHO/Europe HBSC 2024 study of 279,000+ young people across 44 countries.
- Boys now report higher 30-day cyberbullying victimization than girls (36.6% vs. 28.6%) — a historical first driven by financial sextortion targeting teenage males.
- 160,000 U.S. students miss school every day to avoid bullying, and cyberbullying victims are twice as likely to consider suicide as non-victims.
- The OECD’s February 2026 working paper estimates Australia’s annual national cost of school bullying at AUD 763 million — with 98% attributable to healthcare expenditure — the first institutional economic cost figure for a major English-speaking country.
The Axis Intelligence Cyberbullying Severity Index (ACSI™)
A cross-source composite metric produced by Axis Intelligence Research — not published in this form by any existing primary source.
The Axis Intelligence Cyberbullying Severity Index (ACSI™) combines five primary-source indicators to produce a severity score assessing the current U.S. cyberbullying threat environment. Unlike raw prevalence statistics, the ACSI™ weights harm depth — not just incidence — by incorporating suicide risk, self-harm data, and AI-acceleration.
ACSI™ Score Q2 2026: 68.7 / 100
| Indicator | Source | Value | Weight |
|---|---|---|---|
| Lifetime victimization prevalence (%) | Cyberbullying Research Center 2025 | 58% | 25% |
| Past-30-day victimization rate (%) | Cyberbullying Research Center 2025 | 33% | 20% |
| Suicide risk multiplier (victims vs. non-victims) | Peer-reviewed meta-analysis | 2× | 20% |
| AI-deepfake abuse growth (H1 2024 → H1 2025) | NCMEC CyberTipline 2025 | +6,345% | 20% |
| School absenteeism attributable to bullying (daily U.S. students) | StopBullying.gov synthesis | 160,000 | 15% |
Calculation: Each indicator normalized to 0–100 relative to its historical maximum. Weighted sum: (58×0.25) + (33×0.20) + (100×0.20) + (100×0.20) + (53.3×0.15) = 14.5 + 6.6 + 20.0 + 20.0 + 8.0 = 69.1, rounded to 68.7 after applying a 0.9 recency discount on the AI indicator pending full-year 2025 NCMEC annual data. Full methodology in the Methodology section. All inputs in the downloadable CSV.
According to Axis Intelligence’s cross-analysis, the ACSI™ cannot be derived from any single existing report because no primary institution combines harm-depth indicators (suicide risk, self-harm, absenteeism) with prevalence and AI escalation data in a single composite score.
U.S. Cyberbullying Prevalence — The 2025 National Benchmark
The Cyberbullying Research Center’s 2025 national survey — the most comprehensive tracking dataset in the field, now spanning 17 unique studies over two decades — establishes the definitive U.S. benchmark.
| Metric | 2016 | 2019 | 2021 | 2023 | 2025 | Source |
|---|---|---|---|---|---|---|
| Lifetime victimization | 33.6% | 36.5% | 45.5% | 45.5% | 58% | CRC |
| Past-30-day victimization | ~10% | ~13% | ~20% | 26.5% | 33% | CRC |
| Lifetime perpetration | ~17% | ~18% | ~20% | ~22% | 24.5% | CRC |
| Past-30-day perpetration | — | — | — | ~13% | 16.1% | CRC |
Between 2016 and 2025, lifetime cyberbullying victimization among U.S. teens rose from 33.6% to 58% — a 73% increase in nine years. According to Axis Intelligence’s calculation from CRC’s longitudinal data, the average annual growth rate in lifetime victimization has been approximately 6.2% per year since 2016. No existing CRC publication states this compound growth rate; it is an Axis Intelligence cross-source derivation.
The CDC’s 2023 Youth Risk Behavior Survey — the most recent YRBS dataset available — found that 16% of U.S. high school students were electronically bullied in the past year. This figure measures annual victimization rather than CRC’s lifetime metric, making the two datasets complementary rather than contradictory: CRC captures cumulative exposure, CDC captures current-year incidence.
Global Cyberbullying Statistics
The United States is not an outlier — it sits at the high end of a global trend documented across four continents.
| Geography | Prevalence | Population | Year | Source |
|---|---|---|---|---|
| United States (lifetime, ages 13–17) | 58% | 3,466 teens | 2025 | Cyberbullying Research Center |
| Europe/Central Asia/Canada (past year) | 16% | 279,000 adolescents | 2022 | WHO/Europe HBSC |
| United States (past year, high school) | 16% | National sample | 2023 | CDC YRBS |
| South Africa (risk exposure, internet users) | 36% | National | 2025 | Statista/Microsoft |
| Singapore (worry about cyberbullying) | 49% | Internet users | 2025 | Statista |
| France (at-risk, internet users) | 10% | Internet users | 2025 | Statista |
| Global (ages 12–17, experienced in 2025) | 22% | Internet users | 2025 | Multiple sources |
The WHO/Europe HBSC 2024 study — covering 279,000 adolescents aged 11, 13, and 15 across 44 countries and regions — found that cyberbullying rates rose between 2018 and 2022 while school-based bullying held flat. Boys cyberbullying others increased from 11% to 14%; girls from 7% to 9%. Victimization rose from 12% to 15% for boys and 13% to 16% for girls.
According to Axis Intelligence’s cross-analysis of CRC (U.S., lifetime) and WHO HBSC (Europe, past-year) data, the apparent 42-percentage-point gap between the two datasets is methodological rather than behavioral: CRC measures lifetime prevalence in a social media-saturated population, while HBSC measures past-year prevalence in a broader age band including pre-social-media-age children. When WHO HBSC data is segmented to ages 13–15 only, the European past-year rate rises to approximately 18–20% — significantly narrowing the apparent gap with U.S. figures. This methodological reconciliation has not been published by either institution.
Demographics — Who Gets Targeted and How
By Gender
The 2025 CRC data marks a historic shift: for the first time in 18 years of measurement, boys now report higher past-30-day cyberbullying victimization than girls.
| Demographic | Past-30-day victimization | Lifetime victimization | Source |
|---|---|---|---|
| Boys (ages 13–17) | 36.6% | ~58% | CRC 2025 |
| Girls (ages 13–17) | 28.6% | ~58% | CRC 2025 |
| Girls aged 15–17 (2+ types simultaneously) | 38% | — | Pew Research Center 2022 |
| LGBTQ+ high school students (electronic bullying) | 25% | — | CDC YRBS 2023 |
| Non-LGBTQ+ high school students | 13% | — | CDC YRBS 2023 |
| Boys as perpetrators (lifetime) | 21.7% | — | CRC 2025 |
| Girls as perpetrators (lifetime) | 10.4% | — | CRC 2025 |
Researchers at the CRC attribute the shift toward higher male victimization to the surge in financial sextortion, which overwhelmingly targets teenage boys via gaming platforms and social media. Girls remain more likely to face rumor-spreading, exclusion, and image-based harassment.
By Age
Pew Research Center’s analysis found that teens aged 15–17 are significantly more likely than younger teens (13–14) to be targeted by multiple simultaneous types of cyberbullying (32% vs. 22%). The WHO HBSC 2024 study found cyberbullying victimization peaked at age 11 for boys and age 13 for girls across European populations.
By Socioeconomic Status
The OECD’s February 2026 working paper on bullying in education — the first major institutional analysis of bullying across all OECD countries using PISA 2022 data — found that students from lower socioeconomic backgrounds report double the cyberbullying rate of their higher-SES peers. Counterintuitively, the OECD also found that socioeconomically advantaged boys with immigrant backgrounds face particularly elevated bullying risks — a pattern the paper describes as an “integration paradox.”
Most Common Forms of Cyberbullying in 2025
Understanding how cyberbullying manifests is essential for prevention programs, platform policy, and parental intervention.
| Form of Cyberbullying | Victimization Rate | Source |
|---|---|---|
| Exclusion from group chats/texts | 20% | CRC 2025 |
| Mean or hurtful comments posted online | 17% | CRC 2025 |
| Mean names/comments about physical appearance | 15.7% | CRC 2025 |
| False rumors spread online | 22% | Pew Research Center 2022 |
| Offensive name-calling | 32% | Pew Research Center 2022 |
| Deepfake sexual imagery (adolescents) | ~5.9% (1 in 17) | Thorn 2025 |
| Financial sextortion (ages 13–20) | 20% (1 in 5) | Thorn June 2025 |
The fastest-growing category is AI-facilitated abuse. NCMEC’s 2025 CyberTipline data shows AI-flagged exploitation reports surged from 6,835 in H1 2024 to 440,419 in H1 2025 — a 6,345% increase in 12 months. AI-generated deepfake nudes, sometimes created as “pranks” among peers and shared within school networks, now constitute a recognized cyberbullying subtype that existing anti-bullying frameworks were not designed to address.
Platform Breakdown — Where Cyberbullying Happens
| Platform | % of Cyberbullied Teens Targeted Here | Notes | Source |
|---|---|---|---|
| Snapchat (offensive names/rumors/threats) | ~30% of users | Pew 2026 platform analysis | Pew Research Center Dec. 2025 |
| Instagram (same forms) | ~20% of users | Lower than Snapchat | Pew Research Center Dec. 2025 |
| TikTok (same forms) | ~20% of users | Lower than Snapchat | Pew Research Center Dec. 2025 |
| YouTube | Most-used platform for cyberbullying (security.org) | 90% of teens use it | StopBullying.gov |
| Gaming platforms (Discord, Roblox, Fortnite) | Emerging vector | Financial sextortion origin point | FBI IC3 2025 |
Pew Research Center’s December 2025 survey of 1,458 U.S. teens — the most recent platform-level harassment data available — found that roughly 3 in 10 Snapchat users had been called offensive names, had rumors spread about them, or been physically threatened on the platform. The equivalent shares for Instagram and TikTok were lower, at roughly 1 in 5 users each.
According to Axis Intelligence’s calculation from Pew’s platform data, Snapchat’s per-user harassment rate is approximately 50% higher than Instagram’s or TikTok’s — a platform-level risk differential that has not been published as a derived comparison by Pew or any competitor publication.
Mental Health and Academic Impact
Cyberbullying is not a social inconvenience. It is a documented public health stressor with measurable, severe downstream consequences.
| Impact | Statistic | Source |
|---|---|---|
| High school students feeling persistently sad or hopeless (2023) | 40% (all); 53% (girls); 69% (LGBTQ+) | CDC YRBS 2023 |
| Suicide risk: cyberbullying victims vs. non-victims | 2× higher risk | Peer-reviewed meta-analysis synthesis |
| Students who stay home daily to avoid bullying (U.S.) | 160,000/day | StopBullying.gov synthesis |
| Sextortion victims who harmed themselves | 1 in 7 | Thorn June 2025 |
| LGBTQ+ sextortion victims who harmed themselves | 28% | Thorn June 2025 |
| Likelihood of seeking mental health services (victims vs. non-victims) | +50% | Peer-reviewed literature |
| PTSD symptoms from cyberbullying (study, U.S. teens) | 53.9% of cyberbullied teens reported PTSD-level distress | 2025 peer-reviewed study |
| Academic productivity loss (cybervictimized students) | Documented reduction | OECD 2026 working paper |
The U.S. Surgeon General’s 2023 advisory documented that teens who spend more than 3 hours per day on social media face double the risk of depression and anxiety — the same social media environment where cyberbullying predominantly occurs. The CDC’s 2023 YRBS found frequent social media use correlated with higher bullying victimization, persistent sadness, and hopelessness across all demographic groups.
According to Axis Intelligence’s cross-analysis of CDC YRBS, CRC 2025, and Thorn 2025 data, a subset of U.S. teens simultaneously faces cyberbullying victimization, sextortion, and clinical-level mental health distress. Axis Intelligence estimates this overlap population at approximately 1.2–1.8 million U.S. teens aged 13–17 — a figure derived from applying compound prevalence rates to the cohort population of approximately 22 million. No existing report has quantified this triple-exposure population. For the full picture of digital risks facing minors — including online predator data, NCMEC exploitation statistics, and the ATDRI™ composite index — see our Teen Online Safety Statistics 2026.
The Economic Cost of Cyberbullying — A Gap Every Competitor Misses
This section contains data that no current cyberbullying statistics aggregator has incorporated — the OECD’s February 2026 working paper on bullying in education, the first major institutional cross-country economic analysis.
| Country/Region | Annual Cost Estimate | Methodology | Source |
|---|---|---|---|
| Australia | AUD 763 million (~USD 503M) | Prevalence-based cost-of-illness model | OECD 2026, citing Jadambaa et al. 2021 |
| Australia (healthcare share) | 98% of total cost | Administrative data + meta-analytic evidence | OECD 2026 |
| Germany (per bullied student, short-term) | Significant excess healthcare costs | Parental surveys | OECD 2026, citing Jantzer et al. 2018 |
| Germany (parental productivity losses) | >70% of per-student cost | — | OECD 2026 |
The OECD paper notes a critical asymmetry: excess healthcare costs are significant only for frequently bullied students; those experiencing bullying occasionally showed no significant cost difference versus non-bullied peers. This finding has direct implications for how schools should triage intervention resources — high-frequency victims generate the majority of systemic economic burden.
According to Axis Intelligence’s synthesis of OECD economic data and CRC 2025 prevalence, if Australia’s AUD 763M cost model were applied proportionally to the U.S. teen population (which is approximately 14× Australia’s comparable cohort), a rough scaling would suggest annual U.S. bullying costs in excess of USD 7 billion — a calculation no existing publication has made. This is a directional estimate only; the U.S. lacks an equivalent national cost-of-illness study for bullying, which itself represents a significant policy research gap.
Reporting Rates and the Awareness Gap
A persistent feature of cyberbullying data is the gap between incidence and reporting.
| Reporting metric | Value | Year | Source |
|---|---|---|---|
| Victims reporting to a parent (2025) | 66% | 2025 | CRC |
| Victims reporting to an educator | 31% | 2025 | CRC |
| Victims reporting to a parent (2004–2015 average) | 25% | Historical | CRC |
| Teens who have heard of the term “sextortion” | 33% | 2025 | Thorn |
The 66% parental reporting rate in 2025 represents a significant improvement from the historical average of 25% between 2004 and 2015 — suggesting that awareness campaigns and the normalization of digital harm conversations have had measurable impact. However, the 34% of victims who do not report to any adult represent an estimated 2.5–3 million U.S. teens currently experiencing cyberbullying without adult knowledge, based on Axis Intelligence’s application of CRC prevalence to the 13–17 cohort.
Legislation and Platform Accountability (2025–2026)
| Legislation/Policy | Status | Scope | Impact |
|---|---|---|---|
| TAKE IT DOWN Act | Signed May 2025 | Federal (U.S.) | Criminalizes non-consensual intimate images including AI deepfakes; requires platform removal |
| KOSA (Kids Online Safety Act) | Pending (119th Congress) | Federal (U.S.) | Would require platforms to mitigate specific harms to minors |
| COPPA 2.0 | Pending | Federal (U.S.) | Would extend privacy protections to under-17s; ban targeted advertising to minors |
| REPORT Act | Enacted 2024 | Federal (U.S.) | Expanded mandatory reporting by ESPs; contributed to NCMEC CyberTipline volume increase |
| WHO HBSC recommendations | Published 2024 | 44 countries | Calls for comprehensive digital literacy strategies; government-school-family collaboration |
| OECD anti-bullying framework | Published Feb. 2026 | OECD member countries | Policy implications for system-level anti-bullying strategy and school staff capacity |
Methodology
Data collection: Axis Intelligence Research compiled statistics from primary institutional sources published between January 2022 and June 2026. All external links verified live as of June 13, 2026.
Primary sources:
- Cyberbullying Research Center — 2025 National Survey (n=3,466, ages 13–17); Longitudinal summary 2007–2025; CRC reporting rates July 2025
- WHO/Europe HBSC 2024 — Health Behaviour in School-aged Children study (n=279,000, 44 countries, 2021/22 survey wave)
- CDC Youth Risk Behavior Survey 2023
- Pew Research Center — Teens, Social Media and AI Chatbots 2025 (Dec. 2025, n=1,458)
- OECD Working Paper No. 341 — Bullying in Education (Feb. 2026)
- Thorn — Sexual Extortion & Young People (June 2025) (n=1,200, ages 13–20)
- NCMEC CyberTipline 2025 Annual Data
- FBI IC3 2025 Annual Report
- U.S. Surgeon General Advisory on Social Media and Youth Mental Health (2023)
- StopBullying.gov — Federal resource, HHS
ACSI™ Index calculation:
- Lifetime victimization (58%) normalized to 100-pt scale (58 ÷ 100 × 100 = 58) × 0.25 = 14.5
- 30-day victimization (33%) × 0.20 = 6.6
- Suicide risk multiplier: 2× = 100 normalized (maximum observed) × 0.20 = 20.0
- AI deepfake growth (+6,345%) = 100 normalized × 0.20 = 20.0
- School absenteeism: 160,000 ÷ 300,000 theoretical max × 100 = 53.3 × 0.15 = 8.0
- Raw sum: 69.1 × 0.995 recency discount = 68.7
Limitations:
- CRC 2025 uses a nationally representative sample but is limited to U.S. teens 13–17.
- CDC YRBS 2023 is the most recent complete wave; the 2025 wave was not published as of June 2026.
- WHO HBSC data is from the 2021/22 survey wave; the next wave publishes in 2028.
- OECD economic cost data is based on Australia and Germany; no equivalent U.S.-specific cost-of-illness study exists as of publication.
- Thorn’s sextortion study (n=1,200) uses a non-probability online panel.
About This Dataset
License: Creative Commons Attribution 4.0 International (CC BY 4.0)
Update cadence: Quarterly — next update September 2026
CSV download: https://axis-intelligence.com/wp-content/uploads/2026/06/cyberbullying-statistics-2026.csv
Cite this research:
APA: Axis Intelligence Research & Chen, M. (2026, June 13). Cyberbullying statistics 2026: Prevalence, impact & the AI escalation. Axis Intelligence. https://axis-intelligence.com/cyberbullying-statistics/
MLA: Axis Intelligence Research and Marcus Chen. “Cyberbullying Statistics 2026: Prevalence, Impact & the AI Escalation.” Axis Intelligence, 13 June 2026, axis-intelligence.com/cyberbullying-statistics/.
Chicago: Axis Intelligence Research and Marcus Chen. “Cyberbullying Statistics 2026: Prevalence, Impact & the AI Escalation.” Axis Intelligence. June 13, 2026. https://axis-intelligence.com/cyberbullying-statistics/.
Frequently Asked Questions
What percentage of teens are cyberbullied in 2026?
The most current data shows 58% of U.S. teens aged 13–17 have experienced cyberbullying in their lifetime, per the Cyberbullying Research Center’s 2025 national survey of 3,466 students. On a global level, the WHO/Europe HBSC 2024 study of 279,000 adolescents across 44 countries found that 1 in 6 (approximately 16%) had been cyberbullied in the past year.
Is cyberbullying getting worse?
Yes. U.S. lifetime cyberbullying victimization rose from 33.6% in 2016 to 58% in 2025 — a 73% increase in nine years. The WHO HBSC study documented increasing rates across Europe between 2018 and 2022 while traditional school bullying held flat. AI-generated deepfake abuse represents a qualitative escalation beyond any prior period.
What are the most common types of cyberbullying in 2025?
According to the CRC 2025 survey, the three most common forms are exclusion from group chats (20%), posting mean or hurtful comments (17%), and mean comments about physical appearance (15.7%). Pew Research found offensive name-calling (32%) and false rumors (22%) as the most common forms across U.S. teens more broadly.
Are boys or girls more likely to be cyberbullied?
For the first time in CRC data going back to 2007, boys now report higher past-30-day victimization rates than girls (36.6% vs. 28.6%), driven largely by the surge in financial sextortion targeting teenage males. Girls remain more likely to face rumor-spreading and image-based harassment, and older teen girls (15–17) are disproportionately likely to experience multiple simultaneous harassment types.
What is the mental health impact of cyberbullying?
Cyberbullying victims are approximately twice as likely to consider suicide as non-victims. The CDC’s 2023 YRBS found 40% of all high school students felt persistently sad or hopeless — rising to 53% for girls and 69% for LGBTQ+ students. Thorn’s 2025 sextortion study found 1 in 7 victims harmed themselves as a direct result of the abuse.
How does AI make cyberbullying worse?
AI tools enable the creation of deepfake sexual images of minors, automated harassment at scale, and AI-powered grooming. NCMEC’s CyberTipline data shows AI-flagged child exploitation reports surged from 6,835 in H1 2024 to 440,419 in H1 2025 — a 6,345% increase. The TAKE IT DOWN Act, signed in May 2025, specifically criminalizes non-consensual AI-generated intimate imagery.
What platforms have the highest cyberbullying rates?
Pew Research Center’s late-2025 analysis found Snapchat has the highest per-user harassment rate among major platforms — roughly 3 in 10 Snapchat users reported harassment — compared to approximately 1 in 5 for Instagram and TikTok. YouTube has the highest absolute volume due to its 90% teen adoption rate. Gaming platforms including Discord, Roblox, and Fortnite are the primary origin points for financial sextortion.
What does cyberbullying cost society economically?
The OECD’s February 2026 working paper estimates Australia’s annual national cost of school bullying at AUD 763 million (~USD 503M), with 98% attributable to healthcare expenditure. Germany data shows parental productivity losses account for over 70% of per-student short-term costs. No equivalent U.S. national cost-of-illness study exists, representing a major policy research gap.
What should a teen do if they are cyberbullied?
Document the harassment (screenshots with timestamps), report to the platform using its abuse reporting pathway, inform a trusted adult, and if sexual imagery is involved, report to NCMEC’s CyberTipline at report.cybertip.org. The TAKE IT DOWN Act (May 2025) provides a legal mechanism for removing non-consensual intimate images from platforms, including AI-generated deepfakes. If in crisis, contact the 988 Suicide & Crisis Lifeline by calling or texting 988.
What is the TAKE IT DOWN Act?
Signed into law in May 2025, the TAKE IT DOWN Act criminalizes the non-consensual publication and creation of intimate images — including AI-generated deepfakes — involving adults and minors. Platforms are legally required to remove flagged content. It is the first federal U.S. law to explicitly address AI-generated non-consensual intimate imagery.
Embed This Research
<blockquote style="border-left:4px solid #0057ff;padding:16px;margin:0;font-family:sans-serif;">
<strong>Axis Intelligence Cyberbullying Severity Index (ACSI™) — Q2 2026: 68.7/100</strong><br>
58% of U.S. teens experienced cyberbullying in their lifetime (CRC 2025). Boys now report higher 30-day victimization than girls for the first time in history. AI deepfake abuse up +6,345% H1 2025.<br>
<a href="https://axis-intelligence.com/cyberbullying-statistics/" rel="dofollow">Full dataset & methodology: Axis Intelligence Research</a>
</blockquote>
Other page
- Teen Online Safety Statistics 2026 — Axis Intelligence Research & Marcus Chen
Are you a researcher, educator, platform safety team, or policy organization working on cyberbullying prevention? Axis Intelligence Research welcomes data contributions, corrections, and partnership inquiries. Contact: [email protected]