GLOBAL

TRANSPARENCY
REPORT
H1 2024
2 Global Transparency Report | H1 2024

GLOBAL
TRANSPARENCY
REPORT H1 2024

At X, our purpose is to serve the public

conversation, ensuring a safe environment where
everyone can participate freely and confidently. With
this transparency report, we aim to cement X as a
truly safe platform for all. We achieve this through
transparency about our content moderation
systems, policies, and enforcement philosophy. Our
Global Transparency Report covers X Safety
Enforcement for the period of January to June 2024.
3 Global Transparency Report | H1 2024

X RULES ENFORCEMENT

Our mission at X is to promote and protect Our policies and enforcement principles
the public conversation. We believe X are grounded in human rights, and we
users have the right to express their have been taking an extensive and holistic
opinions and ideas without fear of approach towards freedom of expression
censorship. We also believe it is our by investing in developing a broader range
responsibility to keep users on our of remediations, with a particular focus on
platform safe from content that violates education, rehabilitation, and deterrence.
our Rules. Violence, harassment, and These beliefs are the foundation of
other similar types of behavior discourage “Freedom of Speech, not Freedom of
people from expressing themselves, and Reach” - our enforcement philosophy,
ultimately diminish the value of global which means we restrict the reach of
public conversation. posts, only where appropriate, to make the
content less discoverable as an alternative
to removal. The X Rules and policies are
publicly accessible on our Help Center.

When determining whether to take enforcement action, we may consider a number of

factors, including (but not limited to) whether:

‣ The behavior is directed at an individual, group, or protected category of people;

‣ The report has been filed by the target of the abuse or a bystander;
‣ The user has a history of violating our policies;
‣ The severity of the violation; and
‣ The content may be a topic of legitimate public interest.

To enforce our Rules, we use a Our human moderators use important

combination of machine learning and context to make decisions about potential
human review. These systems either take violations. This work is led by an
action automatically, or surface content to international, cross-functional team with
human moderators based on user reports 24-hour coverage and the ability to cover
and/or proactive detection methods. multiple languages. We also have an
appeals process for any potential errors
that may occur.
4 Global Transparency Report | H1 2024

USER REPORTS
AND TOTAL ENFORCEMENT ACTIONS
(EXCLUDING PLATFORM MANIPULATION & SPAM)

We strive to create an environment where users feel

empowered to express themselves. When abusive
behavior occurs, we make it easy for users to report
violations of X Rules or local laws. Below, you'll find
the total number of global user reports from January
to June 2024, excluding Platform Manipulation and
Spam, which is broken out separately below.

USER REPORTS 224,129,805

% OF TOTAL
POLICY AREA TOTAL
REPORTS

Abuse & Harassment 81,730,426 36.47%

Child Safety 8,928,019 3.98%

Hateful Conduct 66,898,539 29.85%

Illegal or Regulated Goods and Services 35,695 0.02%

Misleading & Deceptive Identities 5,179,431 2.31%

Non-Consensual Nudity 38,736 0.02%

Private Content 9,902,566 4.42%

Suicide & Self Harm 2,479,097 1.11%

Violent & Hateful Entities 8,932,100 3.99%

Violent Content 40,005,196 17.85%

5 Global Transparency Report | H1 2024

ENFORCEMENT
ACTIONS
When we take enforcement actions, we may do so
either on a specific piece of content (e.g., an
individual post or Direct Message) or on an account.
Below is a summary table of our automated and
TOTAL ACCOUNT
manual Rule Enforcement Actions and post violation SUSPENSIONS 5,296,870
rates that depict the percentage of all posts that were
actioned for violating one of our platform rules during TOTAL POSTS
REMOVED OR LABELED
10,675,980
the covered time period. This data may include
potential false positives and does not attempt to OVERALL POSTS
account for false negatives. VIOLATION RATE 0.0123%

ACCOUNT POSTS REMOVED POST

POLICY AREA SUSPENSIONS OR LABELED VIOLATION RATE

Abuse & Harassment 1,102,778 2,648,475 0.0030%

Child Safety 2,781,634 14,571 <0.0001%

Hateful Conduct 2,361 4,950,321 0.0057%

Illegal or Regulated Goods & Services 514,095 549,328 0.0006%

Misleading & Deceptive Identities 677,798 62 <0.0001%

Non-Consensual Nudity 52,093 156,498 0.0002%

Private Content 1,719 34,497 <0.0001%

Suicide & Self Harm 2,286 86,184 0.0001%

Violent & Hateful Entities 57,185 153 <0.0001%

Violent Content 104,921 2,235,891 0.0026%

6 Global Transparency Report | H1 2024

POLICY
SPECIFIC
BREAKDOWNS

In the following pages, we break down our

enforcement actions by policy area over a six-month
period from January to June 2024. It is further
broken down into the specific actions taken, and if
they were performed through automated or human
review.
7 Global Transparency Report | H1 2024

CHILD SAFETY

At X, we have zero tolerance for child sexual We send a mix of automated and human
exploitation and are committed to removing reviewed reports to NCMEC. We will never stop
media that depicts physical child abuse. We also using the full extent of our capabilities to
suspend users who engage with that content to eliminate these harmful practices.
prevent the normalization of violence against
children. Along with taking action under our Read our comprehensive policy that outlines how
Rules, we also work closely with the National we make our platform inhospitable for those who
Center for Missing and Exploited Children seek to exploit minors in any way here.
(NCMEC).

TOTAL ACTIONS AUTOMATED HUMAN

NCMEC Reports 370,588 35,176 335,412

Accounts Suspended 1 2,781,634 2,388,683 392,951

Content Removed 14,571 1,645 12,926

1. Accounts suspended for engaging with (liking, replying, bookmarking, etc.) Child Sexual Abuse Media (CSAM) are not reportable to NCMEC

ABUSE, HARASSMENT, & HATEFUL CONDUCT

At X, we are committed to a respectful Additionally, we prohibit ongoing behavior and

community. We prohibit direct attacks based on content that targets others with abuse or
race, ethnicity, national origin, caste, sexual harassment, or encourages other people to do
orientation, gender, gender identity, religious so. Our policies are designed to ensure everyone
affiliation, age, disability, or serious disease. We can engage safely and respectfully, as defined
take swift action against any violations to ensure here.
a safe space for everyone. Read more here.

TOTAL ACTIONS AUTOMATED HUMAN

Accounts Suspended 1,105,139 154 1,104,985

Content Removed 2,247,107 5,679 2,241,428

User Informed Labels 5,351,689 3,966,430 1,385,259

8 Global Transparency Report | H1 2024

PLATFORM MANIPULATION & SPAM

We want X to be a place where people can Multiple accounts that exhibit a pattern or
make human connections, find reliable sole purpose of the above behaviors, or
information, and express themselves freely found to be misusing X product features to
and safely. To make that possible, we do not disrupt others’ experiences, may be
allow spam or other types of platform suspended by X. Posts may not appear in
manipulation. We define Platform certain parts of the X product, and/or may not
Manipulation as using X to engage in bulk, be recommended or amplified by X due to
aggressive, or deceptive activity that multiple product quality or Rule enforcement
misleads others and/or disrupts their reasons. Learn more about the instances
experience. when a Post's visibility may be limited.

TOTAL ACTIONS AUTOMATED HUMAN

Accounts Suspended 463,960,085 2 461,172,954 2,787,131

User Informed Labels 187,689,809 187,255,600 434,209

2. Our Platform Manipulation & Spam defenses are primarily proactive or automated. We remove multiple times more accounts than we receive user reports.

VIOLENT & HATEFUL ENTITIES

At X, we have zero tolerance for threats or individuals promoting violence—to use our
promotion of terrorism and violent extremism. platform. Such content jeopardizes the safety
We do not allow Violent or Hateful Entities— and well-being of targeted individuals.
including terrorist organizations, extremist Read more here.
groups, perpetrators of violent attacks, or

TOTAL ACTIONS AUTOMATED HUMAN

Accounts Suspended 57,185 35,420 21,765

Content Removed 153 62 91

9 Global Transparency Report | H1 2024

VIOLENT CONTENT

X is a place where people can express As a result, we may remove or reduce the
themselves, show and learn about what’s visibility of Violent Content to ensure the
happening, and debate global issues, often safety of our users and prevent the
sharing images and videos as part of the normalization or glorification of violent
conversation. However, healthy conversations actions. We also do not allow sharing Violent
can’t thrive when Violent Speech is used to Content in highly visible places such as
deliver a message, and not every participant profile photos, banners or bio.
wishes to be exposed to Violent Media. Read more here.

TOTAL ACTIONS AUTOMATED HUMAN

Accounts Suspended 104,921 4 104,917

Content Removed 1,797,553 1,103,131 694,422

User Informed Labels 438,338 0 438,338

SUICIDE & SELF HARM

At X, we are dedicated to supporting mental Our policy is designed to provide support and
health and preventing harm. We do not permit resources for those experiencing self-harming
content that promotes or encourages suicide behaviors or suicidal thoughts.
or self-harm. Read more here.

TOTAL ACTIONS AUTOMATED HUMAN

Accounts Suspended 2,286 90 2,196

Content Removed 86,184 58 86,126

10 Global Transparency Report | H1 2024

MISLEADING & DECEPTIVE IDENTITIES

At X, authenticity is crucial. We do not tolerate the

misappropriation of identities or the use of fake identities to
deceive others. Our policies are designed to prevent any form
of identity fraud or misleading behavior on our platform.
Read more here.

TOTAL ACTIONS AUTOMATED HUMAN

Accounts Suspended 677,798 470,435 207,363

Content Removed 62 0 62

NON-CONSENSUAL NUDITY POLICY

At X, sharing explicit sexual images or videos To learn more about our non-consensual
of someone without their consent is strictly nudity policy, read more here.
prohibited, whether the content is AI-
generated or organically created. This severe
violation of privacy, often referred to as
revenge porn, poses serious safety and
security risks and can lead to significant
physical, emotional, and financial hardship for
those affected.

TOTAL ACTIONS AUTOMATED HUMAN

Accounts Suspended 52,093 520 51,573

Content Removed 156,498 22,653 133,845

11 Global Transparency Report | H1 2024

PRIVATE CONTENT

At X, protecting privacy is a priority. You may not threaten,

publish, or share other people's private information or media
without their express authorization and consent. We enforce
strict measures to safeguard personal privacy and prevent
unauthorized disclosures. Read more here.

TOTAL ACTIONS AUTOMATED HUMAN

Accounts Suspended 1,719 4 1,715

Content Removed 34,497 4,047 30,450

ILLEGAL OR REGULATED GOODS/SERVICES

You may not use our service for any unlawful purpose or in
furtherance of illegal activities. This includes selling, buying,
or facilitating transactions in Illegal Goods or Services, as
well as certain types of regulated goods or services.
Read more here.

TOTAL ACTIONS AUTOMATED HUMAN

Accounts Suspended 514,095 84,482 429,613

Content Removed 549,328 132 549,196

12 Global Transparency Report | H1 2024

GOVERNMENT, LEGAL, & LAW

ENFORCEMENT REQUESTS

At X, government, legal, and law enforcement

requests are managed through our established
guidelines. We provide clear procedures for law
enforcement seeking account information and
content removal, and may file or serve objections in
certain circumstances 3. For detailed information and
to learn about requests to withhold content,
please read more.

3. For example, X may le or serve objections for requests that are legally defective, overly broad, and/or appear to impermissibly burden free expression.
fi
13 Global Transparency Report | H1 2024

INFORMATION AND REMOVAL REQUESTS

This section presents the latest data on government requests for

X account information globally, including both emergency and
non-emergency inquiries.

TOTAL NUMBER CASES DISCLOSURE

OF REQUESTS DISCLOSED RATE

TOTAL INFORMATION REQUESTS 18,737 9,897 52.82%

USA 3,329 2,522 76%

Japan 2,726 1,627 60%

EU 7,872 4,388 56%

UK 635 307 48%

All Others 4,175 1,135 27%

TOTAL NUMBER
CASES ACTIONED ACTION RATE
OF REQUESTS

TOTAL REMOVAL REQUESTS 72,703 51,487 70.82%

Turkey 9,364 6,332 68%

Japan 46,648 36,731 79%

South Korea 5,889 4,296 73%

EU 2,458 1,965 80%

All Others 8,345 2,163 26%

14 Global Transparency Report | H1 2024

COPYRIGHT AND TRADEMARK NOTICES

Respecting intellectual property is crucial. Trademark law helps prevent unauthorized

At X, we take intellectual property rights or confusing use of these trademarks.
seriously, including both copyright and
To learn more about how we uphold these
trademark protections. A trademark or
rights and ensure compliance, read our
copyright can be a word, logo, or phrase
detailed Trademark policy here and
that identifies a brand's goods or services.
Copyright policy here.

NOTICES TOTAL URLS WITHHOLDING/

TOTAL
ACTIONED OR ACTIONED OR RESTORATION
NOTICES
WITHHELD WITHHELD RATE

Copyright Notices 188,661 112,897 836,194 59.84%

Copyright Counter Notices 14,496 1,698 1,698 11.71%

Trademark Notices 70,980 6,296 6,296 8.87%

Global Transparency Report | H1 2024