Terms Watch

Removed sections for California and US State users, Brazilian users, policy changes, contact information, and specific data categories collected for marketing communications.

Effective date changed from October 12, 2026 to October 16, 2026. Added provisions allowing users to share data across Google services for personalization and ads, and added an exception for personalized ads based on Drive, Gmail, or Google Photos content.

Effective date changed to 16 October 2026. Added options to let services like Search or Gemini access data from Photos, Gmail, and Drive for personalization and ads, and added a Personal Intelligence section.

Added Prompt-to-Voice service terms requiring express consent to reproduce voices, disclosure of AI-generated speech, preservation of origin indicators, and excluding Prompt-to-Voice and its output from indemnification.

Added certain content of public interest as eligible to earn ad revenue under Harmful Misinformation and Hate Speech/Harassment sections, and added a definition for Public Interest.

Added __Host-ot_bridge essential cookie for security and xai_analytics_early_start non-essential cookie for analytics to the trackers list.

Added sections detailing data collection for Reddit Ads, Pro, and Programs, third-party embeds, data use and sharing practices, user rights, international transfers, and regional disclosures for the EEA, UK, US, and Brazil.

Added the xm-kdt cookie for recognizing browsers that have passed PIN or passkey authentication, updated cookie domains and controllers, and converted cookie details to a table.

Removed the majority of the Privacy Policy sections, including details on information usage, sharing, protection, user rights, international transfers, and jurisdiction-specific disclosures.

Prohibited all gambling ads targeting Brazil. Removed South Korea from permitted countries for smoking alternative ads and removed its standalone vape device terms. Allowed cryptocurrency ads in Hungary.

Added a note stating that account access cannot be provided to anyone regardless of relationship, and added a section detailing the deactivation process and required documentation for incapacitated users.

Effective date changed from August 7, 2026 to October 8, 2026. Added provisions regarding stored credentials in Third-Party Connectors and renamed Third-Party Products to Third-Party Technologies.

Expanded Non-Consensual Nudity policy to prohibit creating, manipulating, or generating intimate AI deepfakes and non-consensual media, added explicit consent exceptions, and updated reporting procedures.

Removed detailed sections on filing copyright complaints, counter-notices, retractions, processing timelines, information forwarding, and repeat infringer suspension rules.

Restored the Safety and Privacy Centre and Platform Rules content, detailing guidelines on dangerous, deceptive, sensitive, and illegal content, reporting processes, and age assurance. (Note: May be tracking system issue)

Effective date changed to October 12, 2026. Added __Host-id-token-nonce and Google Sign-In g_state cookies for grok.com to the list of essential cookies.

Renamed Gemini Spark to Tasks in Gemini (or task mode) and updated references to Search Services History and Timeline settings.

Added a provision referencing terms and support links for New Bundled Subscription Plans, and modified company address details from registered offices to principal place of business.

Effective date changed from 1 October 2026 to 12 October 2026. Clarified collection of data not associated with accounts across devices and updated legal grounds and disclosure examples for sharing data with law enforcement and government authorities.

Effective date changed from October 1, 2026 to October 12, 2026. Added clarifying language on cross-device activity tracking when signed out and expanded legal grounds for disclosing data to law enforcement and in response to legal requests.

Added legal basis breakdowns for data processing under Legitimate Interest, Legal Obligation, and Consent, along with dedicated compliance sections for US and Brazilian users.

Added sections covering international data transfers, compliance with EU-U.S., UK Extension, and Swiss-U.S. Data Privacy Frameworks, and legal bases for EEA, Swiss, and UK users.

Removed the request not to include personal information in prompts. Updated payment data terms to cover general purchases and added user responsibility terms for provided payment information.

Removed sections covering International Data Transfers, Data Privacy Frameworks, Additional Information for EEA, Swiss, UK, California, and Brazilian Users, Policy Changes, and Contact details.

Changed all references to the company and brand name from SpaceXAI to SpaceXSI throughout the guidelines.

Changed company short name reference from SpaceXAI to SpaceXSI throughout the agreement.

Added oauth_link to the list of essential security cookies and changed company references from SpaceXAI to SpaceXSI.

Updated references throughout the document from SpaceXAI to SpaceXSI.

Changed entity references and defined terms from SpaceXAI to SpaceXSI throughout the policy.

Changed entity references and naming throughout the document from SpaceXAI to SpaceXSI.

Removed the previously displayed superseded Terms of Service sections that were effective until October 9, 2026.

Added information categories used for consent-based processing, dedicated privacy disclosures for California and other US states, and a specific section for Brazilian users under the LGPD.

Added detailed legal basis tables detailing information categories used under Legitimate Interest, Legal Obligation, and Consent for various data processing activities.

Added the __Secure-castle_oauth_token_* security cookie with a 100-second duration to the list of essential cookies.

Removed legal bases processing tables for legal obligation and consent, along with sections for California and US state users, Brazilian users, policy changes, and contact information.

Effective date changed from September 29, 2026 to October 7, 2026. Added StackAdapt Ads cookies (user-id, user-id-v2, sa-camp-<ID>, and sa_aid_pv) to the cookie list.

Effective date changed to November 12, 2026. Expanded and reorganized Universal Usage Standards and High-Risk Use Case Requirements, adding specific rules for high-risk physical actions, verification programs, and favorable decision exceptions.

Added legal basis processing tables, state privacy rights disclosures for California and other US states, and a dedicated section for Brazilian users under LGPD.

Removed sections detailing legal bases for data processing, California and US state user disclosures, Brazilian user rights under LGPD, policy change notifications, and contact details.

Added several OpenAI, Cloudflare, and Impact cookies to the policy table, changed the lifespan of oai-did from 1 year to 2 years, and updated __cflb expiration from 30 minutes to 60 minutes.

Added legal bases tables for legal obligations and consent, alongside new sections for California and US state users, Brazilian users, policy changes, and regional contact information.

Removed sections detailing Legal Obligation and Consent bases, Additional Information for California and Other US State Users, Additional Information for Brazilian Users, Changes to This Policy, and Contact Us information.

Updated section header and added text explaining that Gemini can remember specific facts, interests, and preferences across conversations depending on eligibility and settings.

Replaced references to redditors with users across multiple guidelines sections, including spam, impersonation, prohibited goods and services, and developer rules.

Added sections on International Data Transfers, Data Privacy Frameworks disclosures, additional terms for EEA, Swiss, UK, California/US, and Brazilian users, policy changes, and contact information.

Removed sections covering International Data Transfers, Data Privacy Frameworks, regional disclosures for EEA/UK/Brazil/US states, policy changes, and contact details.

Added policy sections detailing how user information is used, shared, and protected, user rights and choices, international data transfers, regional disclosures for EEA/UK/Switzerland, California, and Brazil, and contact details.

Removed sections detailing How We Use Your Information, How We Share Your Information, How We Protect Your Information, Your Rights and Choices, international data transfers, regional disclosures, and contact information.

Expanded guideline scope to explicitly cover generative AI, usernames, display names, and manipulative behaviors. Added specific enforcement actions including removing content, issuing warnings, limiting visibility, account termination, and prohibiting rejoining without permission.

Added consent categories for research and marketing, state-specific disclosures for California and other US states, and a dedicated section for Brazilian users under LGPD.

Effective date changed from April 2, 2026 to October 1, 2026.

Removed sections detailing US state user disclosures, Brazilian user provisions, changes to the policy, and regional contact information.

Effective date changed from 2 April 2026 to 1 October 2026.

Removed the option and link to file policy violation reports for users who do not have a Reddit account across multiple sections.

Added DPF arbitration terms via JAMS and new regional privacy sections for EEA, Swiss, UK, California/US, and Brazilian users detailing legal bases and rights.

Removed dispute resolution details via JAMS, regional privacy disclosures for EEA, Swiss, UK, US state, and Brazilian users, policy change terms, and contact information.

Updated X Corp. address description from registered office to principal place of business.

Expanded the Brazilian users section with LGPD rights, controller details, international data transfer rules, and DPO information. Added standard sections on policy changes and regional contact details.

Added a notice stating updated Commercial Terms will go into effect on October 30, 2026, and updated Meta's dispute notice mailing address to 1 Meta Way.

Removed detailed Brazilian privacy provisions including user rights, international data transfers, DPO contact details, the Changes to This Policy section, and regional contact information.

Expanded scope of agreement to include Qwenstudio and modified prohibited activities to disallow falsely representing AI-generated output as human-generated (previously non-AI-generated).

Added a note directing law enforcement requests for data of users located in the United States to the U.S. version of the TikTok Law Enforcement Guidelines.

Effective date changed to September 30, 2026. Added app update requirements, noted end of support for Cortana Services, increased Rewards redemption limits for Microsoft products to 1,550,000 points, and updated Copilot supplemental terms.

Effective date changed from September 22, 2026 to September 29, 2026. Added several Hightouch analytics cookies and xAI advertising and analytics cookies to the cookie list.

Added terms for Dots and Agentic Features covering user responsibility, oversight, and purchases. Added terms for Sign in with ChatGPT regarding developer integration and user authentication data sharing.

Removed the Spotify Partner Program Monetization Policies section, including content eligibility rules for Premium Video Revenue and policy violation terms.

Removed the section on Violent and graphic content, including its specific examples and exceptions for newsworthy events and awareness-raising.

Added clarification that API access may include ongoing or proactive interactions, updated App Requests scope to include operating the App, and expanded the definition of App Responses to include all content provided in connection with the App.

Added legal bases tables for data processing (Legitimate Interest, Legal Obligation, Consent), expanded data categories for trends monitoring, and added dedicated sections for US state users and Brazilian users (LGPD).

Added the cursor-apple-relay security cookie for accounts.x.ai to the list of essential cookies.

Removed legal basis tables for Legitimate Interest, Legal Obligation, and Consent, as well as sections for California/US state users, Brazilian users, policy changes, and contact information.

Added a minimum age requirement of 13. Expanded community rules into 13 detailed categories, detailed enforcement tiers, and added EU Digital Services Act appeal and dispute procedures.

Restored the Safety and Privacy Centre document after an error page. Added Platform Rules, content actions, age assurance procedures, parent guidance, election integrity, recommendation systems, and DSA transparency sections. (Note: May be tracking system issue)

Effective date changed from August 5, 2026 to September 25, 2026. Legal compliance and security risk updates now take effect immediately. Replaced customer update objection procedures with termination as the sole remedy.

Added a privacy section for the Agentic Calling capability on Pixel devices detailing data sharing, use of transcripts, and storage. Added connected folders to data processed and shared by Gemini Spark.

Effective date changed to September 24, 2026. Added a detailed list of prohibited scam practices, optional policy training for account restoration, and provisions to suspend accounts if a user is at risk of off-platform harm.

Added harassing pranks to the list of protected-against content types.

Added harassing pranks to the list of prohibited content and removed formatting from the identity verification provision.

Effective date changed from September 16, 2026 to September 22, 2026. Added accounts.x.ai to the domain list for the __Host-sso-complete cookie.

Updated eligibility criteria for creator monetization to include active Plus, Premium, Super, or Ultra plans on The X Subscription.

Renamed Harmful or Violent Behavior section header to Harmful, Violent, or Criminal Behavior.

Updated the X Corp. contact address for privacy inquiries from Bastrop, TX to Austin, TX.

Added a provision subjecting the use of credit score services, including connecting to and accessing credit scores, reports, monitoring, and personalized insights, to the Credit Score Terms.

Added functional cookies for age verification and sidebar visibility, introduced an Analytics Cookies section with Google Analytics cookies, renamed Advertising Cookies to Marketing Performance Cookies, and updated the contact email.

Added clarification on Service tier designations when models invoke other models. Added terms governing separately licensed or open-weight models. Expanded Third-Party Features provisions to include third-party identity providers.

Added prohibitions against unauthorized visual tracking or re-identification of individuals, unreviewed output-based decisions affecting fundamental rights, health, or safety, and added biometric info to sensitive data restrictions.

Effective date changed to September 07, 2026. Operating entity changed to NTH POWER GLOBAL TECH SINGAPORE PTE. LTD. Added details on input data processing by Alibaba Cloud and supplementary terms for Brazil, Indonesia, the Philippines, and Singapore.

Contracting entity changed to Nth Power Global Tech Singapore Pte. Ltd. Added subscription and billing terms, model training opt-out provisions, SIAC arbitration, and region-specific terms for Mexico, Brazil, and Indonesia.

Added Hungary to the list of targeted countries for financial advertisements with specific licensing requirements, while excluding it from cryptocurrency and DeFi advertising.

Effective date changed to February 1, 2025. Added arbitration notices, entity definitions, detailed eligibility requirements, a $100 payment threshold, and sections covering taxes, indemnification, invalid activity, termination, and anti-corruption/trade controls.

Document text was updated from English to German, with revised terms on invalid activity definitions, third-party fee deductions, and Spotify Rights Center music claims.

Added legal bases tables for data processing, new state-specific disclosures for California and other US states, and privacy rights sections for Brazilian users.

Mastodon either introduced new Privacy Policy or these terms are being tracked for the first time.

Mastodon either introduced new Terms of Service or these terms are being tracked for the first time.

Removed legal bases tables for processing, jurisdiction-specific sections for California and Brazilian users, changes to policy terms, and contact information.

Effective date changed from September 3, 2026 to September 16, 2026. Added an oauth_fresh_login essential cookie for login on accounts.x.ai with a 2-minute expiration.

Added legal bases tables for Legal Obligation and Consent, state-specific disclosures for California and other US states, and a dedicated section for Brazilian users under LGPD.

Removed sections for California, US State, and Brazilian users, including rights under CCPA and LGPD. Also removed the 'Changes to This Policy' and 'Contact Us' sections.

The effective date was removed. The Third-Party Services section was expanded to include additional services and features. The Termination section now allows for subscription or account cancellation and changed the inactivity period from over a year to more than 120 days.

Replaced "Rules" with "Community Standards" throughout the document. Added an in-app flow option for updating phone numbers. Expanded the content moderation section with details on automated tools, human review, and specific provisions for EU users under the Digital Services Act.

The effective date changed from July 8, 2026 to September 10, 2026. Added supplemental disclosures for residents of the Republic of Korea, including details on data collection, processing delegation, and overseas data transfers.

Added South Korea to the list of countries where smoking alternative ads are permitted and added specific restrictions for advertising in South Korea.

Added a policy stating Instagram may require identity verification to use certain products and features, and may restrict or disable accounts if verification is not completed.

Added a policy stating that identity verification may be required to use certain products and features, and failure to comply may result in restricted access or a disabled account.

Added new OpenAI cookies for service functionality and new Snapchat cookies for marketing measurement.

Added a new section for Financial Plugins and ChatGPT for Financial Services, stating they are for informational purposes only and that OpenAI does not provide financial advice.

Added new terms effective October 9, 2026. The license you grant X now explicitly includes using your content to train AI models. Added liquidated damages for scraping and a new limitations period for filing claims. Updated X Corp.'s registered office address.

Renamed the "Fraud, Scams, and Deceptive Practices" section to "Prohibited Commercial Practices".

Renamed the "Fraud, Scams, and Deceptive Practices" section to "Prohibited Commercial Practices" and added a new "Prohibited Documents and Financial Services" section under "Restricted Goods and Services".

Removed the "Unacceptable Business Practices" section. Added a "Prohibited Documents and Financial Services" section restricting ads for fake documents, money muling, and money laundering.

Renamed the "Fraud, Scams, and Deceptive Practices" policy to "Prohibited Commercial Practices" and rewrote its rationale. Added a new "Prohibited Documents and Financial Services" section to the Restricted Goods and Services policy, detailing rules on fake documents, money laundering, and more.

Expanded device and usage data collection to include interactions with websites and apps. Removed Copilot from the list of services available to child accounts. Updated details on how Copilot uses conversation history for personalization and removed references to Microsoft 365 Copilot.

Added sections on how information is used, shared, and protected. Added details on user rights, international data transfers, and specific provisions for users in the EEA, UK, Switzerland, California, other US states, and Brazil.

Added a new tracker named 'x-workload-version' used for skew protection during service deploys.

Added detailed tables explaining data use by legal basis (Contract, Legitimate Interest, Legal Obligation, Consent). Added new sections with specific rights and information for users in California, other US states, and Brazil.

Removed detailed tables explaining the legal basis for data processing, including Performance of Contract, Legitimate Interest, Legal Obligation, and Consent. Also removed sections with additional information for users in California, other US states, and Brazil.

Added sections for California, other US states, and Brazilian users detailing their specific privacy rights. Added a 'Consent' section explaining how data is used for personalization, advertising, and research. Added 'Changes to This Policy' and 'Contact Us' sections.

Removed the Consent processing basis table, sections for California and US state users, Brazilian users, changes to the policy, and contact details.

The effective date changed to September 3, 2026. The "Memory" feature is now called "Knowledge Base," and it will not seek to store sensitive data included in user input.

Added a class action waiver to the governing law section. Updated arbitration rules to specify that claims can arise before the agreement existed, require disclosure of arbitration funding, and clarify that Spotify can also opt cases out of mass arbitration.

Updated monetization rules for violent content, changing the time window for graphic gameplay violence from 15 seconds to 7 seconds. Modified policies for content featuring recreational drugs and drug trade organizations.

Added a warning that joking about engaging in severe harm may result in account actions or disabling, regardless of intent. Changed section header from 'Threats, Violence & Harm' to 'Threats, Violence, and Harm'.

Added legal obligation and consent processing tables, regional privacy disclosures for US and Brazilian users, policy update terms, and contact details for US, EEA, UK, and Brazil entities.

Effective date changed from August 26, 2026 to September 3, 2026. Added console.x.ai to the domains using the gb_anon_id cookie.

Removed sections detailing legal bases and consent for information processing, state-specific disclosures for California and other US users, Brazilian user information under LGPD, changes to the policy, and regional contact information.

Added links for Digital Services Act and Terrorist Content Online reports. Removed a section explaining the Terrorist Content Online Transparency Report and links to the 2024 and 2025 reports.

Added sections with specific privacy information and rights for users in California, other US states, and Brazil. Added a section on policy changes and updated contact information.

Added a rule restricting content that depicts, promotes, or encourages participation in mid-risk viral challenges to users 18 and over.

Added a restriction for content depicting, promoting, or encouraging participation in mid-risk viral challenges, limiting it to viewers 18 and older.

Added a comprehensive usage policy outlining scope, principles, prohibited content and activities, enforcement, and reporting procedures. Prohibited activities include illegal acts, CSAM, hate speech, misinformation, and security violations.

A comprehensive Data Processing Addendum was added, outlining definitions, roles, obligations for data processing, security measures, subprocessors, international data transfers, audits, and specific terms for CCPA.

Added comprehensive Commercial Terms of Service, including sections on product access, use restrictions, data ownership, confidentiality, indemnification, liability, fees, privacy, and mobile application terms.

Added a comprehensive Cookie Policy detailing the use of technically required, analytics, and marketing cookies from providers like Cloudflare, Google, Posthog, X, and Meta. The policy also explains how users can manage their cookie preferences.

The document was populated with a full privacy policy, including details on data collection, usage, retention, sharing, user rights, and specific disclosures for U.S. users. (Note: May be tracking system issue)

Added comprehensive terms for EEA consumers, including sections on subscriptions, data usage, third-party connectors, mobile apps, and country-specific terms for France, Germany, Italy, Poland, and Belgium.

Updated the effective date from November 1, 2025 to September 1, 2026. Removed sections on arbitration, minimum eligibility, payments, taxes, indemnity, invalid activity, and anti-corruption, deferring to the new Snap Creator Payment Terms.

Removed sections detailing rights for users in California, other US states, and Brazil. Also removed the 'Changes to This Policy' and 'Contact Us' sections.

Added a notice announcing updated Advertising Guidelines effective November 12, 2026, and updated formatting and phrasing in the energy drinks disclaimer requirement for Armenia.

Added notice of updated Community Guidelines effective 12 November 2026. Added counterfeit products to the list of harmful or deceptive practices and products that are removed or restricted.

Added notice of an upcoming Privacy Policy update effective 12 November 2026 with a preview link.

Added notice of updated Terms of Service effective 12 November 2026 and changed onsite and offsite to on-site and off-site.

The effective date changed to September 1, 2026. Added dispute resolution terms requiring Texas courts or arbitration and a class action waiver. Added company address and updated terms for Beta Modes, AI-generated disclosures, and the limitations period for state law claims.

Added sections covering International Data Transfers, Data Privacy Framework compliance, legal bases for EEA/UK/Swiss users, US state disclosures, and Brazilian LGPD compliance.

Removed sections on International Data Transfers, EU/UK/Swiss Data Privacy Frameworks, and additional information for users in the EEA, UK, Switzerland, California, other US states, and Brazil. Also removed the 'Changes to This Policy' and 'Contact Us' sections.

Added a rule against imagery of physical fights outside of a sports context. Added rules for fan, satire, and parody accounts, requiring them to disclose their unofficial status.

Removed references to 'Automated Data Collection Terms' and the requirement to comply with them when collecting information in an automated way.

Added a new section on fan, satire, and parody accounts, requiring them to clearly disclose their unofficial status to avoid restriction or removal.

Added terms for a "Coding Harness Subscription" to access services, including details on fees, usage allowances, and how disputes are handled for subscribers.

Added a rule prohibiting imagery of physical fights between two or more people, except in a sports context.

Added details on user rights under Brazil's LGPD, international data transfers, and contact information for Data Protection Officers in Brazil, the EEA, and the UK.

Removed sections detailing LGPD rights for Brazilian users, international data transfers, the Data Protection Officer, policy changes, and contact information.

The effective date changed from August 24, 2026 to August 26, 2026. The cookie list was expanded to include new essential, analytics, marketing, and fraud detection cookies from xAI, OneTrust, Castle, Mixpanel, and Braze.

Added a summary for the Sora discontinuation, covering content exports, data deletion, refunds, credits, and the API shutdown.

Added a notice that the Community Guidelines will be updated on September 24, 2026.

The company name referenced in the guidelines was changed from 'xAI' to 'SpaceXAI' throughout the document.

The company name has been updated from xAI to SpaceXAI throughout the policy.

The company name has been updated from X.AI LLC to SpaceXAI LLC throughout the policy. The effective date changed from April 4, 2026 to August 24, 2026.

The effective date changed from June 26, 2026 to August 24, 2026. The company name was updated from xAI to SpaceXAI throughout the document.

Added collection of ad history and interests for Free and Go users. Added receiving information from advertisers and data partners to measure ad effectiveness. Added a new purpose for using personal data to personalize ads for Free and Go users. Expanded the legal basis table with details on processing for personalized and generic ads.

Added information on exercising privacy rights, CCPA metrics, and a new section for Brazilian users detailing rights under LGPD, the data controller, international data transfers, and DPO contact information. Also added sections on policy changes and contact details.

The effective date changed from August 12, 2026 to August 24, 2026. Added new cookies for Google Ads, Twitter (X), TikTok Ads, LinkedIn Ads, Microsoft Ads, Reddit Ads, and Snap Ads. Expanded the domains for several existing cookies.

Removed sections on exercising rights, CCPA metrics, Brazilian user rights under LGPD, policy changes, and contact information for the US, EEA, UK, and Brazil.

Added detailed legal basis tables for EEA, Swiss, and UK users, alongside dedicated sections covering rights and disclosures for California, US states, and Brazilian users.

Removed detailed tables on legal bases for data processing for EEA/Swiss/UK users. Also removed sections with additional information for users in California, other US states, and Brazil, as well as the 'Changes to This Policy' and 'Contact Us' sections.

Added a requirement for users in Brazil between the minimum age and the age of legal majority to obtain a judicial permit to produce monetized or sponsored content.

Added clarification that services are for commercial use unless specified otherwise. Expanded the definition of Inputs and added restrictions against selling or transferring API keys. Added a clause prohibiting reselling or providing access to services on a standalone basis.

Added details on how advertising is displayed in public channels, bots, stories, and search results. Advertising can now be targeted based on language, approximate location (country/city from IP or phone number), and the topics of channels a user follows.

Updated country-specific advertising guidelines, changing rules for ads from "Shopping Ads" to all ads. Added or removed prohibitions and restrictions for numerous categories including alcohol, healthcare, financial services, and pharmacies in many countries.

The effective date changed from November 3, 2025 to August 21, 2026. Added collection of ID, birth year, and location for account recovery. Added collection of government ID name and phone number for identity verification. Added age enforcement as a use of data.

Added extensive details on information collection, including for non-account holders, interactions with AI features, and data from partners. The policy was restructured with new sections and examples for clarity on data use, sharing, and user controls.

Removed the contact phone number from the 'How to Contact Us' section.

The full Community Guidelines have been published, replacing a 404 error page. The new guidelines have an effective date of September 13, 2025 and cover topics including safety, content moderation, authenticity, and account enforcement. (Note: May be tracking system issue)

Added information for Brazilian users under the LGPD, including rights, international data transfers, and a Data Protection Officer. Added details on exercising rights via an authorized agent and a link to CCPA metrics. Added contact information for users in Brazil.

Removed sections on exercising privacy rights, CCPA metrics, additional information for Brazilian users, policy changes, and contact information for the US, EEA, UK, and Brazil.

Added clarification distinguishing suspension duration (24 hours to 30 days) from violation expiration, specifying that the Appeals Portal expiration date indicates when a violation is removed from an account record.

Updated the contact email for policy exceptions from aup-exceptions@meta.com to aup_exceptions@meta.com.

In the list of Connected Apps available on Android mobile devices, "Utilities" was changed to "Device assistance".

The title changed to 'Supplemental Terms of Use' and the effective date was updated to August 18, 2026. The terms now apply to the Copilot app and Copilot Chat in Microsoft 365, but not for work or school accounts. A Microsoft account is now required to use Copilot.

Added a banner about updates to the consumer Privacy Notice and removed the "Last updated" date from the document.

Added a Liability section, updated subprocessor notification methods, replaced the static subprocessor table with an external Trust Center link, and changed input data retention from the later to the earlier of deletion or 30 days post-termination.

Added introductory notice about Privacy Notice updates. Changed physical mailing address for copyright notices and contact from 115 Sansome St. to 2261 Market Street, Ste. 14441.

The legal entity changed from X.AI LLC to SpaceXAI LLC. The governing law changed from Tennessee to Texas, and the venue changed from Wichita/Tarrant County to exclusively Wichita County. The limitations period for claims was extended from one year to two years and one day.

The company name changed from xAI to SpaceXAI throughout the document. The effective date was updated to August 14, 2026. The restriction on developing competing services was expanded to include assisting others and competing indirectly.

Added Google marketing measurement cookies GBRAID and WBRAID.

Removed a notice that a new policy will take effect on February 16, 2026.

Added a hotline for Colorado law enforcement personnel to submit inquiries, with responses provided via email to an authenticated law enforcement address.

Updated how Twitch uses personal information to include testing and improving technology like machine learning and AI, and to develop or deploy generative AI models and services.

Added a requirement for users in Brazil between the minimum age and the age of majority to obtain a judicial permit for monetized content. Expanded the list of entities not eligible to use monetization tools to include politicians, campaign committees, and government entities.

The effective date changed from August 11, 2026 to August 12, 2026. The domains for Google (GA4) cookies were updated from '.grok.com' to '.grok.com, x.ai, accounts.x.ai'.

Updated the effective date from August 10, 2026 to August 11, 2026. Added a new way to adjust cookie preferences on x.ai by clicking 'Privacy choices' in the footer.

The effective date changed from July 2, 2026 to August 10, 2026. The domains for Meta Ads, Google Ads, and Twitter (X) cookies were expanded to include x.ai and console.x.ai.

Added a section explaining data handling for the 'Speak to Window' feature on Gemini for macOS, which captures audio and the active window to process instructions.

Added a clause stating the English language version of the guidelines is the version of reference and prevails in case of discrepancies with translations.

The effective date changed to August 5, 2026. The definition of Customer Data was expanded. The policy on using data for training was updated, including for Labs and Preview Models. Breach of confidentiality was removed as an exception to the limitation of liability.

Updated the 'Death & tragedy' section for content that can earn ad revenue, allowing for the direct display of the moment of death in certain educational, documentary, or journalistic contexts.

Added Malaysia as a server location. Expanded data collection for security and anti-fraud purposes, including SIM card, mobile carrier, and WiFi information. Updated data sharing practices with related group companies and data transfer mechanisms for EEA/UK users.

The effective date changed to September 21, 2026. Added that Snap may require age verification and may lock or delete underage accounts. Updated content moderation, safety, and dispute resolution sections, including changes to arbitration fees.

Expanded the definition of Services to include Meta Materials and prohibited uses to include Outputs, Actions, and Integrated Products. Added and expanded prohibitions related to fundamental rights, cybersecurity, professional practices, and the operation of AI agents.

Introduced tiers for services: "Standard Services" and "Discounted Services". Content from Discounted Services may be used for training, while content from Standard Services will not. Added terms for "Downloadable Materials" and a new Exhibit B for EU/UK consumer rights.

The effective date changed from September 21, 2026 to April 7, 2025. The age requirement clause was updated, and sections on content moderation and arbitration fees were modified.

Effective date changed to August 7, 2026. Mistral AI Studio and API access are now excluded from EEA consumer terms, a Third-Party Products section was added, and data training exception criteria were updated.

Updated the location of X Corp. from Bastrop, Texas to Austin, Texas, including the mailing address for legal policy matters.

The effective date changed to 21 September 2026. Added that information may be used to estimate, infer, or verify age. Added that automated tools may be used to detect violations and confirm age-appropriate use.

Updated the effective date to September 21, 2026. Added language about age verification and responding to EU governmental requests for user information. Modified sections on content moderation, appeals, and arbitration fees.

Updated gambling ad requirements for Nigeria to specify authorization by the relevant Gaming Authority instead of Nigerian Gambling and Gaming Authorities.

Added United Arab Emirates to the list of approved countries for targeting in the Middle East.

X either introduced new Advertising Content Policy or these terms are being tracked for the first time.

Twitch either introduced new Advertising Content Policy or these terms are being tracked for the first time.

Added that Microsoft may share age group information with Microsoft apps, services, and third-party apps. Added that the Sticky Notes screenshot feature can capture content from other active app windows. Removed a paragraph about refreshing the Privacy Statement.

The effective date changed from 22 May 2024 to 30 July 2026. Added a section on user responsibility for network costs. Removed the section on handling requests for your data. Modified language around legally required updates and commercial guarantees.

Added 'ads.openai.com' as a domain for numerous marketing performance cookies from LinkedIn, Google, Reddit, and Meta. Changed the browser link for 'Explorer' to 'Edge'.

Removed South Korea as a storage location for Japanese users' personal data. The 'Date of last update' was changed from July 31, 2025 to July 31, 2026.

Added reference to Automated Data Collection Terms and requirement to comply with them, and specified that permission for automated access or data collection may be restricted or revoked at any time.

Restructured destination requirements into new subsections covering profile details, functionality, style, and language. Added a new section on keyword requirements for ad placement.

The effective date changed from March 12, 2026 to July 27, 2026. The conditions for opting out of AI model training were modified.

The effective date changed from April 8, 2026 to July 27, 2026. The product name "Le Chat" was changed to "Vibe". The definition of "Input" was expanded to include data accessed through integrations or connected services.

Published new Community Guidelines covering content standards, harassment, hate speech, misinformation, AI-generated media, privacy, illegal activities, and enforcement. The age requirement is set to 16 years old.

Added a new privacy policy. It details information collected (account info, usage data, content), how it's used (to operate the service, for safety), and when it's shared (with vendors, for legal requests). The minimum age is 16.

The previous landing page has been replaced with a full Terms of Service document. The terms establish an age requirement of 16, grant UpScrolled a license to user content, outline moderation policies, and include a mandatory arbitration agreement for US and Canadian users.

Added a section for EU Member State authorities on reporting under the Regulation Addressing Dissemination of Terrorist Content Online (EU 2021/784), including submission procedures and criteria for terrorist content.

Updated the policy for content that will receive no ad earnings, modifying the rules around displaying the moment of death and implied severe bodily harm.

Added a note stating that exceptions may be made for some betting content, such as from Google-certified providers during live sporting events, which may be allowed for all audiences.

Removed "gender identity" and "age" from the list of sensitive personal characteristics for targeting purposes.

Updated the term 'kids' to 'children' in the section on managed accounts.

Expanded restrictions on weight loss content to include weight gain. Added rules against misleading health practices, including claims to cure certain diseases and clickbait tactics. Removed a section on deceptive and misleading practices.

The effective date changed from December 16, 2025 to July 23, 2026. The policy was updated to clarify how information is used for features that are part of Meta AI, replacing previous mentions of "AI at Meta".

In the Health and Wellness section, 'Weight Loss and Cosmetic Products and Procedures' was removed as a category. The restriction on ads for weight loss products was expanded to include 'weight gain'. A rule against generating negative self-perception was replaced with a general prohibition on certain claims.

Expanded the 'Health and Wellness' policy to include 'weight gain' products and services. Added rules against misleading health practices, including claims to cure certain diseases and clickbait tactics. Removed the 'Deceptive and Misleading Practices' section.

The effective date changed from December 16, 2025 to July 23, 2026. Added a section on how information is used to personalize features that are part of AI at Meta. Updated language around AI features and data sharing with third-party AI partners.

Updated the effective date to July 22, 2026 and changed the example of services where cookies are used from 'Le Chat' to 'Vibe'.

Added a notice stating more detail was added about cookies, first-party advertising measurement, and privacy choices. The notice also clarifies that personal data is not sold and queries are not sent to advertisers.

Updated product name from NotebookLM to Gemini Notebook. Removed a question mark from a URL.

Added a clause stating personal information may be shared with TT Commerce & Global Services LLC. Updated the required disclosure for endorsements from 'TikTok Creativity Program Beta' to 'TikTok Creator Rewards Program'.

Removed five Swoogo analytics cookies and added one OpenAI analytics cookie named '__obi'.

Clarified that users in the European Union may contact the Irish Data Protection Commission to raise concerns about data processing practices.

Updated language regarding data sharing, changing from when you use "agentic features" to when you "ask Gemini to do things for you".

Updated the data retention policy for ChatGPT Business, changing from "workspace admins control how long your workspace data is retained" to "workspace admins can control how long your data is retained".

Expanded the 'Under 13 Experience' to include users who are 13 years old in Florida. The policy now refers to 'users of this experience' instead of 'Children'.

Added the 'aws-waf-token' cookie. Removed the '__cf_bm', 'cf_clearance', 'intercom-device-id-*', and 'intercom-session-*' cookies.

Updated the firearms policy to specify that direct sales or links to retail sites, platforms, marketplaces, or peer-to-peer forums selling firearms or certain accessories are prohibited.

Added a requirement for developers whose 'Sign In with X' usage exceeds normal levels for their tier to upgrade to an Enterprise plan.

Updated the 'Our commitment to young listeners' section, adding 'Project RockIT' as a child safety organization. Added details on managed accounts and a new section on age-appropriate experiences for features like music videos and messages.

Added policies for 'Unsatisfying or Off-putting Content' and 'AI Personas Related to Sensitive Topics'. The 'Inauthentic Content' section was renamed to 'Generic or Repetitive Content' and its guidelines were updated, including moving some examples to 'Reused content'.

The document was replaced with a comprehensive set of Community Guidelines, including an effective date of September 13, 2025. New sections detail rules on safety, civility, mental health, mature themes, authenticity, regulated goods, privacy, and content moderation.

Reddit either introduced new Developer Terms or these terms are being tracked for the first time.

Updated Google Pay data controls to reference Google Wallet settings and website. Added a clause stating that user recordings are not used for AI model training, except to address abuse or harm.

Vibe either introduced new Acceptable Use Policy or these terms are being tracked for the first time.

Vibe either introduced new Data Processor Agreement or these terms are being tracked for the first time.

Vibe either introduced new Trackers Policy or these terms are being tracked for the first time.

Vibe either introduced new Commercial Terms or these terms are being tracked for the first time.

Vibe either introduced new Privacy Policy or these terms are being tracked for the first time.

Vibe either introduced new Terms of Service or these terms are being tracked for the first time.

Modified the condition for acting on a DMCA notification from 'valid' to 'complete'.

Expanded the definition of "Apps" to include "plugins" and specified that plugins may contain Apps SDK apps and custom skills.

Updated the data protection representative and Data Protection Officer (DPO) for the EU and UK from Perplexity AI and TrustKeith Ltd to VeraSafe, and added a direct link for data privacy requests.

Added an FAQ section explaining that external brand deal links in Shorts may be removed if they fail an automated safety check or are flagged for a policy violation.

Added prohibitions on creating adult sexual imagery, cloning voices/likenesses without consent, and specific biometric surveillance practices. Added rules for operating AI agents and a process for requesting exceptions to the policy.

The service was renamed from Llama API to Meta Model API. New sections were added for Paid and Unpaid Services, with different content usage rights for each. The age requirement was raised from the age of majority to 18. New restrictions were added, including a prohibition on using outputs to train competing models.

Expanded the description of how models are trained, including adding a new section detailing the stages of training. Updated the sources of personal data for model training, adding details on ClaudeBot, third-party datasets, and internally generated data.

Added a provision referencing information on handling California Consumer Privacy Act (CCPA) requests, and removed the table of contents navigation links.

Effective date changed from April 4, 2026 to July 2, 2026. Added a Non-Essential Cookies classification grouping performance and targeting cookies, and added a Type column to the cookie list table.

Effective date changed from April 10, 2026 to June 26, 2026. Added provisions regarding Agentic Actions, business domain account linking, AI-generated disclosures, and expanded limitation of liability terms.

Effective date changed to June 26, 2026. Added detailed prohibitions on reverse engineering, jailbreaking, metadata removal, nudifying real persons, automated decision-making, and competing AI development.

Added trackers lsid, loauth_state, loauth_verifier, and loauth_return for domains studio.x.com and localizer.x.com, related to Google OAuth sign-in and CSRF protection.

Updated the recipient for DMCA Counter Notifications from '[us/our Designated Copyright Agent]' to 'us'.

Added email address to account creation data. Added inference of likely gender from profile information for ad targeting. Added receiving hashed phone numbers from advertisers for ad matching. The data controller's registered office address was updated.

Added new policies for using music in Facebook Live broadcasts, including minimizing duration and requiring rights holder approval. Added a list of content formats that cannot be monetized, such as static videos and engagement bait.

Removed the table of TikTok entities and user location criteria, detailed examples of user data and TikTok Shop data types, full answers to FAQs, and the list of countries and TikTok Shop entities in Appendix A.

The policy was restructured and rewritten, changing from a numbered list format to a table-based format detailing data collection contexts and purposes. The effective date was removed and replaced with a 'Last updated' date. Sections on the Comet browser were added.

The effective date changed to August 17, 2026. Added provisions for parent-managed accounts for users under 13 and clarified that children can manage their own accounts when they reach the minimum age.

The effective date changed to July 2, 2026. Added information collection for guest users, usernames, age, and parent supervision. Added a digital age verification provider and expanded on information shared with Meta for backups and subscription features.

Effective date changed to July 1, 2026. Added that users may have different feature experiences if they share less information. Added that public content can be reshared by others. Added that information may be collected from app store APIs. Added a new section for Brazilian users.

Removed the 'Streaks' in-app purchase section and added a 'No Ads' subscription, currently for US users. Modified the account suspension policy from a 3-strike rule to a more general one, and changed the company's registered address.

Removed the deprecated section on 'Products or Services in Violation of Trade Policies'.

Removed detailed explanatory sections, definitions, and extensive lists of allowed, not allowed, and FYF ineligible content across all policy areas, reorganizing into a high-level overview format.

Updated policy on high-risk viral challenges to remove all imagery depicting them regardless of context, eliminating previous exceptions that applied sensitivity labels for awareness raising or condemnation.

Updated policy on high-risk viral challenges to remove all imagery depicting them regardless of context, eliminating the previous exception for awareness raising or condemnation and removing the associated warning label policy.

The effective date changed from June 12, 2026 to July 1, 2026. The privacy request metrics were updated for the 2025 calendar year, modifying the number of requests received, complied with, and denied, along with average response times.

Added Gemini app on macOS to the list of Gemini Apps. Added that the beta Gemini Spark feature on macOS can interact with local files in connected folders and will lose access if turned off.

Added that data from third-party services like Model Context Protocol (MCP) server tools may be collected. Added a warning about connecting custom third-party MCP servers. Added information about using temporary chats to control data use for model improvement.

Updated creator monetization requirements to allow connecting an X Money account or a verified Stripe account.

Updated creator subscription requirements to specify followers with Premium, Premium Business, or Premium Organizations subscriptions. Added rules requiring disclosure for AI-generated and synthetic video game depictions of armed conflict.

Updated data processor for identity verification data from Au10tix to Persona.

Updated Suicide and Self-harm policy to prohibit depicting or engaging in self-harm, dangerous stunts, and substance abuse. Daily post limit changed from 2,400 per day to 50 original posts and 200 replies for unverified accounts.

Updated the data processor for identity verification data from Au10tix to Persona.

Updated references from twitter.com to x.com and added a new tracker entry for the x_studio_context cookie associated with studio.x.com.

Expanded harassment, violent content, and hate policies with violation examples and non-violations. Added specific rules covering bots, AI agents, scraping, and app registration requirements.

The document was translated from German to English. Processing locations were updated from country names in German (e.g., 'USA', 'Dänemark') to English (e.g., 'United States', 'Denmark').

Removed sora.chatgpt.com and videos.openai.com from several cookie domains. Changed the domain for Swoogo cookies from devday.openai.com to events.openai.com. Added a Google cookie for service functionality and removed two Google Analytics cookies.

Workspace admins can now view, access, export, and delete end user conversations. Admins, not end users, control data retention. Data may be retained if reasonably necessary to protect services or third parties from harm.

Removed the data control for personalizing ads. Modified the example for notifying parents/guardians of teen users from notifying authorities to just notifying parents/guardians.

Added sections for Licensed Materials, ChatGPT Sites, and OpenAI Academy. The OpenAI Academy section notes that activity may be shared with an organization's administrators if an associated account is used.

Changed the targeting restriction terminology for political ads, CBD products, and cannabis ads from zip code to postal code.

Expanded the Harmful or Violent Behavior section to specify child labour, mistreatment, or abuse under minor safety risks, and added a requirement that conduct comply with local laws and regulations.

Added a requirement for creators to disclose when content is generated or meaningfully altered with AI, clarifying that impersonation policies still apply.

The policy was substantially rewritten and reorganized for clarity. New sections were added for Artificial Intelligence/Copilot, Storage and Processing, Advertising, and Diagnostic Data. The previous U.S. State Data Privacy section was removed and integrated elsewhere.

Added a rule against using AI to copy a person's voice or likeness for impersonation. Added 'Views from signed-out users' to the list of features that may be turned off for content depicting minors.

Added a notice that as of 22 May 2024, these Generative AI Additional Terms of Service no longer apply and are superseded by the main Google Terms of Service, unless for business partners with specific agreements.

Updated terminology for ad revenue states, changing from 'may earn limited or no ad revenue' to 'will receive limited ad earnings' or 'will receive no ad earnings' across multiple policy sections.

Added a notice of a future Terms of Service update effective 30 July 2026. The definition of 'commercial guarantee' was modified.

Added a sentence with a link to a summary of the terms.

The effective date changed to June 11, 2026. The product 'Le Chat' was renamed to 'Vibe'. The method for reporting violations was updated from email to a web form.

Added rules for Direct Messages, Stories, Lounges, and Shop/Commerce. New prohibited content categories include non-consensual intimate imagery and synthetic/manipulated media. The policy for permanent user removal was expanded and the web-only filter policy was updated.

The governing entity changed from Parler Technologies, Inc. to Sovren Technologies, Inc. Added new sections for Direct Messages, Stories, Lounges, Shop/Commerce, Guest Access, SMS Invitations, and Account Deletion. Expanded policies on Boosted Posts, Tipping, and CSAE.

Company name changed to Sovren Technologies, Inc. Added sections on data collection for Direct Messages, Stories, and Lounges. Expanded user rights for state and EEA residents, including rights to know, delete, and opt-out. Age requirement changed from 18 to 13.

Added a link to a summary of the terms and removed the 'Other ways to get help' section.

Changed the name of the linked policy from "Spam, deceptive practices & scams policies" to "Spam policy".

The effective date changed to June 12, 2026. Added a 'Third-Party Links' section. Added collection of Precise Location Data and Customer Service Data. Updated the definition of tailored advertising and how data is used for it. Added data disclosures to advertising partners.

Updated the contact email for regulatory inquiries and for the representative in Korea from dataprotection@wechat.com to dataprotection@corp.wechat.com.

Added a Prohibited Targeting section banning the use of data identifying or inferring sensitive personal characteristics, including race, health, religion, and gender identity, for ad targeting purposes.

Effective date changed to July 8, 2026. Added collection of Study Participation and Verification Data, which may include biometric data. Expanded how Inputs, Outputs, and third-party integrations are handled. Added research publications as a data disclosure category.

The effective date changed from March 26, 2026 to June 8, 2026. Added information on data collection for optional subscription features, such as Meta One, including Device and Connection Information and User Choices.

Removed the 'Documentation Index' section and its associated link from multiple developer documents.

Updated the account deletion process. Data will now be kept for 30 days before permanent deletion, changing from being effective immediately.

Added a rule for AI-generated content made with generic templates. The policy now specifies that the substance of each video should be 'materially varied' instead of 'relatively varied' and changed references from 'ad revenue' to 'ad earnings'.

Updated the Impersonation policy to include using AI to copy a voice or likeness. The Spam policy was restructured and expanded to prohibit engagement manipulation, fake engagement, detection evasion, automated mass-production, and malicious clickbait.

Removed the effective date of 30 May 2026 and replaced it with a 'Last updated' date of 30 May 2026. Removed reference to the previous policy posted on 12 September 2025.

Removed the effective date of 30 May 2026 and a reference to the previous terms. Added a 'Last updated' date of 30 May 2026.

Added a provision allowing the restriction or disabling of accounts belonging to users under 18 that persistently post or share age-inappropriate content, including adult nudity, drugs, alcohol, tobacco, graphic violence, or self-harm.

Added provision allowing the restriction or disabling of accounts belonging to users under 18 that persistently violate Community Standards by sharing content that is not age-appropriate.

Updated the list of platform features, changing from 'Create, Bursts, Connect, DMs, OwnProfile' to 'Find, Search, Watch, Play, Buy, Shop'. Also changed 'Feed, Lounges, and Bursts' to 'Feed, Lounges, Play, and Shop'.

Added collection of Transaction Data for in-app purchases and UX Interaction Recordings for a sample of users. Clarified that interests are inferred from visual elements of photos for ad targeting. Added sections on SMS communications and expanded the U.S. Privacy Rights section.

The effective date changed from April 8, 2026 to May 28, 2026. The product name "Le Chat" was replaced with "Vibe" throughout the document, including in the use restrictions and data training sections.

The effective date changed from March 25, 2026 to May 28, 2026. The product name 'Le Chat' was changed to 'Vibe'. The Third-Party Services section was updated to expand permissions granted to Mistral AI when connecting third-party services.

The company name was changed from "Mistral AI" to "Mistral" in the legal notice.

The effective date changed from March 31, 2026 to July 1, 2026. For users in the UK, the contracting entity changed from Reddit Netherlands B.V. to Reddit, Inc. The address for Reddit Netherlands B.V. and the Copyright Agent were also updated.

Removed Sora content creation and distribution guidelines, replacing them with discontinuation details, data export instructions, data deletion timelines, and refund information.

Added descriptive summaries for the Developer Agreement, Developer Policy, Ads API Agreement, Developer PPU Agreement, Restricted Uses, Geo guidelines, and Display requirements sections.

Modified the newsletter signup form to request a full name in addition to an email address.

Added a link to a new 'Parler Play' privacy policy.

Updated a reference to indemnification obligations from Section 8 to Section 9. A typo was introduced in the disclaimer regarding AI output and intellectual property law.

Added a rule prohibiting content that knowingly uses a service to artificially inflate live traffic, also known as 'viewbotting'.

Added that Meta may retain user content when its systems flag the data for a potential policy violation.

Renamed 'Gemini Agent' to 'Gemini Spark' and expanded its data usage to include a remote computer. Added sections on data use for Avatars and Google shopping cart. Added 'Skills' as a type of instruction provided by users and clarified data collection for tasks.

Added collection of voice, video, and biometric data. Added data collection for personalized services via a Memory feature. The age for data protection was changed from under 13 to under 18. Expanded the purposes for data processing and disclosure.

Added a restriction against removing or modifying labels or watermarks on outputs. Expanded the definition of prohibited abusive content. Added a requirement to disclose when content is AI-generated before sharing. Expanded user responsibility and indemnification clauses related to outputs. Added a clause allowing Qwen to assign or novate the terms.

Added Perplexity Max and Perplexity Health to the policy. Updated data collection to include health information with consent, information from invited users, and expanded on sensitive data collection for US residents. Modified contact details for the DPO and representatives.

Updated the 'No Subscription Refunds' policy to include exceptions for country-specific billing and consumer information pages.

Added Performance and Targeting cookies, including from Meta, Google, Twitter (X), and TikTok. Previously, the policy stated only necessary cookies were used. New controls for managing cookie preferences were also added. The effective date changed to April 4, 2026.