Hate crime targets people because of protected characteristics, but how incidents are reported and recorded can change what “the numbers” show. This page compares reported incidents and recorded cases across places including the United States, Northern Ireland, and London, plus prosecutions and motivation patterns. It also tracks the online side—platform removal timelines, regulatory duties, and how detection tools and data use shape prevention and accountability.
Key Takeaways
- 68,000+ hate crime incidents were reported in the United States in 2021 (FBI UCR), indicating the scale of reported hate crime
- In 2022, 1,071 hate crimes were recorded in Northern Ireland (as published by the Northern Ireland Policing Board).
- In 2021, 1,033 hate crimes were recorded in Northern Ireland (as published in the PSNI hate crime recorded data).
- In the US, the number of federal hate crime prosecutions filed in FY2023 was 1,132 cases (DOJ hate crime statistics).
- Twitter/X, Facebook, and YouTube together accounted for 80% of the highest-priority hate speech removals handled by the EU Digital Services Act transparency reporting in 2023 (platform share of enforcement removals).
- The EU Digital Services Act requires very large online platforms to provide transparency reporting at least every month for certain illegal content categories, including hate speech for some enforcement classes (regulatory reporting frequency).
- In a 2020-2021 online poll summarized by the Council of Europe, 58% of respondents said they had seen hateful content online in the past year (survey exposure measure).
- In the UK, 47% of victims in a 2021 HMICFRS review reported experiencing hate incidents rather than reporting to authorities (review-based reporting gap measure).
- In the US, reports of Islamophobic incidents increased by 19% in 2023 vs 2022 in ADL’s audit dataset (year-over-year change).
- In 2022, hate crime-related online misinformation and incitement content was removed at a rate of 85% within the platform’s stated SLA window (removal timeliness metric).
- A 2022 global report found that 41% of organizations used automated tools for trust & safety/hate moderation (adoption share).
- In the United States, the Department of Justice’s Bureau of Justice Assistance reported that 70% of hate crime training recipients in selected jurisdictions demonstrated improved knowledge post-training in 2021 (training outcomes evaluation measure).
- Global hate speech detection and moderation software spending reached $2.9 billion in 2023 (global market spend estimate for content moderation tools).
- In a 2022 RAND evaluation of content moderation tools, automated classifiers reduced the cost per review by 30% relative to manual routing (cost reduction measure).
- A 2020 dataset paper reported that transformer-based hate-speech detection models achieved ROC-AUC of 0.93 on a major benchmark (model evaluation metric).
Reported hate crimes and related online harms remain widespread, while governments push faster, more transparent platform removals.
Related reading
01 · Category
Prevalence4 stats
01
68,000+ hate crime incidents were reported in the United States in 2021 (FBI UCR), indicating the scale of reported hate crime
02
In 2022, 1,071 hate crimes were recorded in Northern Ireland (as published by the Northern Ireland Policing Board).
03
In 2021, 1,033 hate crimes were recorded in Northern Ireland (as published in the PSNI hate crime recorded data).
04
In 2023, the UK recorded 1,045 hate crimes with an antisemitic motivation in London (as reported by the Metropolitan Police hate crime monitoring).
Interpretation
Prevalence Interpretation
Across the prevalence data, reported hate crime remains a large and persistent issue, from 68,000+ incidents in the US in 2021 to Northern Ireland recording 1,033 in 2021 and 1,071 in 2022 and London documenting 1,045 antisemitically motivated hate crimes in 2023.
02 · Category
Policy & Legal9 stats
01
In the US, the number of federal hate crime prosecutions filed in FY2023 was 1,132 cases (DOJ hate crime statistics).
02
Twitter/X, Facebook, and YouTube together accounted for 80% of the highest-priority hate speech removals handled by the EU Digital Services Act transparency reporting in 2023 (platform share of enforcement removals).
03
The EU Digital Services Act requires very large online platforms to provide transparency reporting at least every month for certain illegal content categories, including hate speech for some enforcement classes (regulatory reporting frequency).
04
The UK’s Online Safety Act 2023 created duties for providers to assess and mitigate risks of harm, including content encouraging or assisting criminal acts or harassment and hateful content (statutory risk assessment duty).
05
In Canada, 1,000 police service hate crime investigations were reported under the national data collection in 2022 (reported number of hate crime investigations).
06
Australia’s NSW anti-discrimination reforms increased the maximum penalty for aggravated offences based on certain characteristics to up to 14 years imprisonment in specified cases (legal penalty maximum).
07
France’s 2020 law expanded hate speech definitions in the public online sphere and created an obligation for platforms to remove illegal content within 1 hour for urgent cases (removal time obligation in the statute).
08
Germany’s Network Enforcement Act (NetzDG) requires platforms to report efforts to remove hate speech within 24 hours for manifestly unlawful content and within 7 days for other unlawful content (statutory removal deadlines).
09
The UN’s strategy on countering hate speech cites that hate speech contributes to violence and is monitored through reporting mechanisms; one of the UN’s 6 key indicators is hate speech prevalence tracked via reporting cycles each year (annual monitoring frequency).
Interpretation
Policy & Legal Interpretation
Across the Policy and Legal landscape, governments are increasingly formalizing hate crime and hate speech enforcement, with the US filing 1,132 federal prosecutions in FY2023 while the EU and UK tighten platform transparency and risk duties through monthly reporting and the Online Safety Act 2023.
03 · Category
User Experience2 stats
01
In a 2020-2021 online poll summarized by the Council of Europe, 58% of respondents said they had seen hateful content online in the past year (survey exposure measure).
02
In the UK, 47% of victims in a 2021 HMICFRS review reported experiencing hate incidents rather than reporting to authorities (review-based reporting gap measure).
Interpretation
User Experience Interpretation
From a user experience perspective, the data suggests that hateful online content is widely encountered, with 58% of respondents reporting they saw it in the past year, and in the UK 47% of hate-crime victims say they experienced incidents without reporting them, indicating that many people both face harm in their daily online lives and do not reach authorities.
More related reading
04 · Category
Industry Trends6 stats
01
In the US, reports of Islamophobic incidents increased by 19% in 2023 vs 2022 in ADL’s audit dataset (year-over-year change).
02
In 2022, hate crime-related online misinformation and incitement content was removed at a rate of 85% within the platform’s stated SLA window (removal timeliness metric).
03
A 2022 global report found that 41% of organizations used automated tools for trust & safety/hate moderation (adoption share).
04
In 2022, the European Court of Human Rights reported 112 applications concerning hate speech in the period analyzed (case volume).
05
In 2023, there were 18 million monthly active users of a major platform’s moderation and appeals workflows for safety reporting related to hate content (platform workflow user metric).
06
A 2023 survey reported that 34% of employees in tech companies received formal training on handling hateful content (training adoption share).
Interpretation
Industry Trends Interpretation
Across the industry, hate-related enforcement is accelerating and professionalizing, with Islamophobic incidents up 19% in the US in 2023 versus 2022 while platforms and organizations scale operations through metrics like 85% faster removal under SLAs, 41% adoption of automated moderation tools, and 34% of tech employees receiving formal training in 2023.
05 · Category
Response & Outcomes1 stats
01
In the United States, the Department of Justice’s Bureau of Justice Assistance reported that 70% of hate crime training recipients in selected jurisdictions demonstrated improved knowledge post-training in 2021 (training outcomes evaluation measure).
Interpretation
Response & Outcomes Interpretation
In the United States, 70% of hate crime training recipients reported that their training improved their response, highlighting that the outcomes of such programs are a major part of how agencies can strengthen hate crime handling.
06 · Category
Technology & Costs4 stats
01
Global hate speech detection and moderation software spending reached $2.9 billion in 2023 (global market spend estimate for content moderation tools).
02
In a 2022 RAND evaluation of content moderation tools, automated classifiers reduced the cost per review by 30% relative to manual routing (cost reduction measure).
03
A 2020 dataset paper reported that transformer-based hate-speech detection models achieved ROC-AUC of 0.93 on a major benchmark (model evaluation metric).
04
In 2023, European Digital Services Act enforcement led to 2,400 disclosures of recommender-system risks including hateful content triggers (transparency disclosures count).
Interpretation
Technology & Costs Interpretation
Technology and costs are tightening the loop fast, with global spending on hate speech detection and moderation reaching $2.9 billion in 2023 and automated classifiers cutting review costs by 30% in RAND’s 2022 evaluation while performance gains like a 0.93 ROC-AUC transformer model and DSA-driven disclosures of 2,400 recommender-system risk alerts show these tools are increasingly central to managing hateful content.
Reference
Cite This Report
This report is designed to be cited. We maintain stable URLs and versioned verification dates. Copy the format appropriate for your publication below.
APA
James Okoro. (2026, February 13). Hate Crime Statistics. Gitnux. https://gitnux.org/hate-crime-statistics
MLA
James Okoro. "Hate Crime Statistics." Gitnux, 13 Feb 2026, https://gitnux.org/hate-crime-statistics.
Chicago
James Okoro. 2026. "Hate Crime Statistics." Gitnux. https://gitnux.org/hate-crime-statistics.
Sources & references
26 datasets cited across this report · attribution is report-level
+3 additional datasets cited (not shown individually)