Separate AI terms govern users living in the UK, Brazil or the European Region
The page opens by saying Meta’s AI terms differ by the region you live in. If you live in the United Kingdom, Brazil or the European Region, a separate regional Meta AIs Terms of Service (linked, not reproduced here) governs your use of Meta’s generative AI products and features instead. The page does not state which other regions the terms below cover; it only lists these three carve-outs.
Supporting quotations
Our AI Terms are different depending on the region you live in
govern the use of our Generative AI products and features if you live in the United Kingdom
govern the use of our Generative AI products and features if you live in Brazil
govern the use of our Generative AI products and features if you live in the European Region
Qualifications & exceptions
The regional pointer is keyed to where you live, not where you are using the product.
The three regional terms are linked documents and are not part of this source.
Not stated here
The UK, Brazil and EU Meta AIs Terms of Service are linked but not supplied; their content is unknown here.
Which regions the main terms below apply to is not stated on the page.
These Terms bind you to Meta Platforms, Inc. when you use "AIs" through any Meta Product
The Meta AIs Terms of Service apply when you access and use Meta AIs and the other software, tools, services, documentation, content, functionalities and generative AI features Meta makes available through any Meta Product; the terms call all of this "AIs". The agreement is between you and Meta Platforms, Inc. ("Meta"). Using AIs counts as agreeing to these Terms, to the separate Meta Terms of Service, and to any other applicable terms and policies.
Supporting quotations
Meta AIs Terms of Service
apply when you access and use Meta AIs and other software, tools, services, documentation, content, functionalities, and generative AI features that we make available to you through any
(collectively, “AIs”). These terms are a legally binding agreement between you and Meta Platforms, Inc. (“Meta”, “we”, “our”, or “us”). By using AIs, you agree to be bound by these Terms, the
and any other applicable terms and policies
Qualifications & exceptions
"AIs" is defined broadly: it covers software, tools, services, documentation, content, functionalities and generative AI features made available through any Meta Product, not only the Meta AI assistant.
Agreement is by use: no separate acceptance step is described.
Not stated here
The Meta Terms of Service and the list of Meta Products are linked documents and are not supplied here.
"Any other applicable terms and policies" is open-ended; which ones apply to a given use is not stated.
Account & Service Control · Restriction or exception
No access to AIs under age 13 (or a higher local minimum) or if the law bars you from Meta services
You cannot access AIs if you are under 13, or under whatever greater age your country or territory requires, or if applicable law prohibits you from receiving Meta’s services or software. This is stated as additional to, not instead of, the eligibility rules in the Meta Terms of Service and other applicable terms or policies.
Supporting quotations
Eligibility
you cannot access AIs if you are under the age of 13 (or such greater age required in your country or territory) or if you are prohibited from receiving our services or software under applicable laws
Without limiting the
Qualifications & exceptions
The minimum age rises to any greater age required in your country or territory.
The rule is "without limiting" the Meta Terms of Service and other policies, so further eligibility limits may exist in those documents.
Not stated here
Any additional eligibility rules in the Meta Terms of Service are not supplied here.
Legal Rights & Agreement Rules · Restriction or exception
You may not use AIs, or let others use them, in any way on the Acceptable Uses list; breach can cost access
The Acceptable Uses section is a list of things you may not do with AIs. The ban covers your own access and use and also allowing others to access or use AIs in any of the listed ways. The list items (explained in the following cards) describe manners of use "that would" produce the listed result, so the test is the effect of the use, not only intent. Elsewhere the terms say that if Meta determines you breached these Terms or other applicable terms and policies, it may suspend or permanently disable your access to AIs or other Meta services, and may permanently disable or delete your account.
Supporting quotations
Acceptable Uses
You may not access, use, or allow others to access or use AIs in any manner that would:
If we determine that you have breached these Terms or other applicable terms and policies, we may suspend or permanently disable your access to AIs or other Meta services or features, and we may permanently disable or delete your account
Qualifications & exceptions
Allowing others to use AIs in a prohibited way is itself prohibited.
The consequence for breach (suspension, disabling, account deletion) is Meta’s determination and is discretionary ("may").
Legal Rights & Agreement Rules · Restriction or exception
Prohibited: uses that violate law, including anything involving child sexual abuse material
You may not use AIs in a way that would violate any applicable law. The terms spell out that this includes promoting, contributing to or encouraging illegal or unlawful activity, and conduct that describes, encourages, supports or distributes child sexual abuse materials in any form.
Supporting quotations
You may not access, use, or allow others to access or use AIs in any manner that would:
Violate any applicable law (including but not limited to any manner that promotes, contributes to, or encourages illegal or unlawful activity), including conduct that describes, encourages, supports, or distributes in any form child sexual abuse materials
Qualifications & exceptions
"Applicable law" is not limited to any one country.
Legal Rights & Agreement Rules · Restriction or exception
Prohibited: content that may harm people, manipulative techniques, or exploiting vulnerable people
You may not generate, promote, disseminate, use or facilitate content that may harm other individuals, including encouraging self-harm, discrimination, bullying, humiliation or other emotional harms. The terms also count as harm using AIs to deploy subliminal, manipulative or deceptive techniques that distort behaviour by impairing decision making, or to exploit people because of their age, disability or a specific social or economic situation in order to distort their behaviour.
Supporting quotations
You may not access, use, or allow others to access or use AIs in any manner that would:
Generate, promote, disseminate, or otherwise use or facilitate content that may harm other individuals, including to encourage self-harm, discrimination, bullying, humiliation, and/or other emotional harms
(i) deploy subliminal, manipulative, or deceptive techniques to distort behavior by impairing decision making; or (ii) exploit people due to their age, disability, or a specific social or economic situation to distort their behavior
Qualifications & exceptions
The threshold is content that "may" harm, not content shown to have harmed.
Inference & Profiling · Restriction or exception
Prohibited: harmful social scoring of individuals and predictive policing based on profiling
You may not use AIs to evaluate or classify individuals based on social behaviour or personal traits where that leads to detrimental or unfavourable treatment, either in a context unrelated to the data’s original context or where the treatment is unjustified or disproportionate. You also may not use AIs for predictive policing or criminal risk assessments based on profiling or on assessing personality traits and characteristics.
Supporting quotations
You may not access, use, or allow others to access or use AIs in any manner that would:
Evaluate or classify individuals based on social behavior or personal traits that leads to detrimental or unfavorable treatment, either in a context unrelated to the original data or that is unjustified or disproportionate
Engage in predictive policing or criminal risk assessments based on profiling or assessing personality traits and characteristics
Qualifications & exceptions
The social-scoring ban is tied to detrimental or unfavourable treatment that is out of context, unjustified or disproportionate; evaluation without that treatment is not addressed.
Inference & Profiling · Restriction or exception
Prohibited: facial-image scraping, biometric categorisation, and emotion or biometric ID uses (two law carve-outs)
Four biometric uses are banned. You may not use AIs for untargeted scraping of facial images from the internet or CCTV footage to build facial recognition databases, or to categorise people by their biometric data to deduce or infer race, political opinions, trade union membership, religious or philosophical beliefs, sex life or sexual orientation. Emotion recognition in the workplace or in education, and real-time biometric identification in publicly accessible spaces for law-enforcement purposes, are also banned, but each of those two carries the words "save as permitted by applicable law", so a legally permitted instance is not prohibited by this list.
Supporting quotations
You may not access, use, or allow others to access or use AIs in any manner that would:
Engage in the untargeted scraping of facial images from the internet or CCTV footage to build facial recognition databases
Engage in emotion recognition in the workplace or education, save as permitted by applicable law
Categorise people based on their biometric data to deduce or infer their race, political opinions, trade union membership, religious or philosophical beliefs, sex life or sexual orientation
Engage in real-time biometric identification in publicly accessible spaces for the purposes of law enforcement, save as permitted by applicable law
Qualifications & exceptions
The "save as permitted by applicable law" exception attaches only to emotion recognition (workplace or education) and to real-time biometric identification for law enforcement; the facial-scraping and biometric-categorisation bans carry no such carve-out.
Legal Rights & Agreement Rules · Restriction or exception
Prohibited: uses presenting a risk of death or bodily harm, and adult or sexual content
You may not use AIs in a manner that would present a risk of death or bodily harm to individuals, which the terms illustrate with illegal drugs, weapons and self-harm. Separately, you may not generate, promote, disseminate, use or facilitate adult content, with erotic chat, pornography and content meant to arouse sexual excitement given as examples.
Supporting quotations
You may not access, use, or allow others to access or use AIs in any manner that would:
Present a risk of death or bodily harm to individuals (including in relation to illegal drugs, weapons, or self-harm)
Generate, promote, disseminate, or otherwise use or facilitate adult content, such as erotic chat, pornography, and content meant to arouse sexual excitement
Qualifications & exceptions
Both items use "including" or "such as", so the named examples are illustrative, not the full set.
Legal Rights & Agreement Rules · Restriction or exception
Prohibited: deceiving or misleading others, including scams, phishing, malware, plagiarism and misinformation
You may not use AIs in a way that would deceive or mislead others. The terms name plagiarism, scams, phishing or malware, and furthering misinformation or disinformation as examples, while saying the list is not exhaustive.
Supporting quotations
You may not access, use, or allow others to access or use AIs in any manner that would:
Deceive or mislead others, including but not limited to activity related to plagiarism, scams, phishing or malware, and furtherance of misinformation or disinformation
Qualifications & exceptions
"Including but not limited to": the named examples are illustrative.
Content & IP Rights · Restriction or exception
Prohibited: content or Prompts that may infringe others’ rights, including intellectual property and privacy
You may not generate, promote, disseminate, use or facilitate content that may violate or infringe the rights of others, with intellectual property rights and privacy rights named as examples. The Rights to Content section repeats the point from the input side: you may not provide Prompts or use AIs in any way that infringes or violates someone else’s rights, again naming privacy and intellectual property.
Supporting quotations
You may not access, use, or allow others to access or use AIs in any manner that would:
Generate, promote, disseminate, or otherwise use or facilitate content that may violate or infringe upon the rights of others (including but not limited to intellectual property rights and privacy rights)
You may not provide Prompts or use AIs in any way that infringes or violates someone else’s rights, including but not limited to privacy and intellectual property rights.
Qualifications & exceptions
The threshold is content that "may" violate or infringe, not a finding of infringement.
The rule applies to what you put in (Prompts) as well as what is produced.
Legal Rights & Agreement Rules · Restriction or exception
Prohibited: seeking professional advice or content for regulated activities; outputs are not professional care
You may not use AIs to solicit professional advice, with medical, psychological, financial and legal advice given as examples, or to obtain content to be used for other regulated activities, such as political campaigning or lobbying. The same item states that no statement, response or content generated by AIs should be read as an offer or provision of professional care or as a substitute for consulting a qualified provider.
Supporting quotations
You may not access, use, or allow others to access or use AIs in any manner that would:
Solicit professional advice (including but not limited to medical, psychological, financial, or legal advice) or content to be used for the purpose of engaging in other regulated activities (including but not limited to political campaigning or lobbying).
No statements, responses, or content generated by AIs should be interpreted as an offer or provision of professional care or as a substitute for consultation with a qualified provider.
Qualifications & exceptions
The named advice fields and regulated activities are examples ("including but not limited to").
The warning section later repeats, in capitals, that you should not use Outputs to inform professional advice or decisions related to medicine, finance, law or pharmaceuticals.
Security & Safeguards · Restriction or exception
Prohibited: overriding safety or privacy filters, controls or mechanisms
You may not use AIs in a manner that would override safety or privacy filters, controls or mechanisms. The item is unqualified: it does not distinguish Meta’s mechanisms from others’ or name any permitted testing.
Supporting quotations
You may not access, use, or allow others to access or use AIs in any manner that would:
Override safety or privacy filters, controls, or mechanisms
Not stated here
Whose filters, controls or mechanisms are meant is not stated.
Content & IP Rights · Context
Defined terms: Prompts, Outputs, Feedback and Content
The terms define four words used throughout. "Prompts" are the materials you provide for processing, which depending on the integration may include text, documents, images or recordings. "Outputs" are the responses AIs generate based on your Prompts. "Feedback" is any feedback you provide. "Content" is the collective term for Prompts, Outputs and Feedback. Later sections about Meta’s use of Content, accuracy disclaimers and responsibility for Outputs rely on these definitions.
Supporting quotations
AIs allows you to provide materials for processing, which could include text, documents, images, or recordings, depending on the integration (together, “Prompts”) and will generate a response (“Outputs”) based on your Prompts, as well as any feedback you provide (“Feedback”) collectively, “Content”.
Qualifications & exceptions
Which input types are available depends on the integration you are using.
You keep the rights you have in your Prompts, except as these Terms expressly provide
Except where these Terms expressly say otherwise, you retain all right, title and interest, including any intellectual property rights, that you have in your Prompts. The exception matters: a later section says Meta uses Content (which includes Prompts) and related information as described in the Meta Terms of Service and Meta Privacy Policy, including to provide, maintain and improve its services, so retaining ownership does not mean Meta makes no use of what you submit.
Supporting quotations
Except as expressly provided in these Terms, you retain all right, title, and interest – including any intellectual property rights – that you have in and to your Prompts.
Meta uses Content and related information as described in the Meta Terms of Service and Meta Privacy Policy
Qualifications & exceptions
Retention is of rights "that you have"; the terms do not grant you rights you lacked.
Meta’s own use of Content under the Meta Terms of Service and Privacy Policy is the express carve-out that limits this statement.
Not stated here
The Meta Terms of Service and Meta Privacy Policy, which describe the extent of Meta’s use, are not supplied here.
Ownership of Outputs is not addressed in this sentence.
Legal Rights & Agreement Rules · Restriction or exception
You promise your Prompts comply with these Terms and other policies, which Meta may revise at any time
You agree that providing Prompts will not violate these Terms, including the Acceptable Uses list, and the other policies that apply to your use of AIs. The same sentence says Meta may supplement, update or revise those other policies at any time. For these Terms themselves, the Changes section says Meta will, unless the law requires otherwise, notify you of material changes (for example by email or in-app notification) and give you a chance to review them before they take effect; continuing to use AIs after an update means you are bound by it.
Supporting quotations
You agree that your provision of Prompts will not violate these Terms, including the Acceptable Uses, and other policies that apply to your use of AIs, which we may supplement, update, or revise, at any time.
Unless otherwise required by law, we will notify you (for example through email or in-app notification) of any material changes to these Terms and give you an opportunity to review them before they go into effect. Once any updated Terms are in effect you will be bound by them if you continue to use our AIs.
Qualifications & exceptions
The any-time revision language is attached to the "other policies that apply to your use of AIs"; material changes to these Terms come with advance notice and a review opportunity, unless law requires otherwise.
If you do not agree to updated Terms you should stop using AIs before they take effect.
Not stated here
Which other policies apply to your use of AIs, and how changes to them are communicated, is not stated.
Use & Processing · Company practice
Where available, media you provide (images, audio, video) is converted into an embedding each time
Depending on where you are located, you may be able to interact with AIs by providing media such as images, audio or video, for example speaking to Meta AI, providing media for editing, or asking questions about media. The terms explain the mechanism: each time you provide media, the AI technology transforms and represents it as an embedding in order to understand the media and deliver the requested service. Media you provide falls within the defined term "Prompts", which covers images and recordings depending on the integration.
Supporting quotations
Depending on where you are located, you may have the option to engage with AIs by providing media (e.g., images, audio, video, etc.)
To work, each time you provide media, the AI technology transforms and represents the media into an embedding to understand the media and provide the requested service.
AIs allows you to provide materials for processing, which could include text, documents, images, or recordings, depending on the integration (together, “Prompts”)
Qualifications & exceptions
Media input is available only in some locations ("Depending on where you are located").
The embedding is described as necessary for the feature to work ("To work"); the terms do not say how long embeddings are kept.
Not stated here
Which locations support media input is not stated.
Retention of embeddings is not addressed here.
Legal Rights & Agreement Rules · Restriction or exception
You warrant you have consent from anyone depicted in media you submit for processing
By providing media to AIs for processing, you represent, warrant and agree that you have, and that you provide to Meta, any appropriate permissions or consents from the people depicted in that media for the processing to take place. This sits alongside the general rule that you may not provide Prompts that infringe or violate someone else’s rights, including privacy rights.
Supporting quotations
By providing media to AIs for processing, you represent, warrant and agree that you have, and otherwise provide to Meta, any appropriate permission(s) or consents from persons depicted in such media to conduct such processing.
You may not provide Prompts or use AIs in any way that infringes or violates someone else’s rights, including but not limited to privacy and intellectual property rights.
Qualifications & exceptions
The warranty is triggered by the act of providing media; no separate confirmation step is described.
It covers permissions from the persons depicted, and the words "otherwise provide to Meta" extend those permissions to Meta.
Not stated here
What counts as "appropriate" permission or consent is not defined.
Your AI interactions are used to personalise experiences and ads and to improve AI at Meta, per the Privacy Policy
The Meta Privacy Policy (linked, not supplied here) is where Meta explains how personal information is processed in the context of AIs. The terms summarise that this includes using your interactions with AIs to personalise your experiences and your ads and to improve AI at Meta. "Interactions" is read broadly: the information you or others exchange with AIs, such as Content or messages, plus the information Meta uses to make responses more relevant, with your interests, location and profile given as examples.
Supporting quotations
Personal Information
explains how we process personal information in the context of AIs, including how we use your interactions with AIs to personalize your experiences and ads, and improve AI at Meta
When you or others interact with AIs, those interactions include information you or others exchange with AIs like Content or messages, along with information we use to make responses from AIs more relevant (like your interests, location and profile).
Qualifications & exceptions
Interactions by "others" with AIs are included, not only your own.
The relevance signals named (interests, location, profile) are examples, not a closed list.
Not stated here
The Meta Privacy Policy, which holds the full description of this processing, is not supplied here.
Retention & Deletion · Company practice
AIs sometimes retain and use what you share; deleting messages or your account may not delete Meta’s copy
When information is shared with AIs, the AIs will sometimes retain and use it. Meta therefore tells you not to share information you do not want the AIs to use and retain, giving information about sensitive topics as the example. The Personal Information section adds a practical warning: depending on how you use AIs, deleting individual messages, entire threads or even your account may not delete Meta’s copy of your personal information, and the Privacy Center describes the correct way to delete it.
Supporting quotations
When information is shared with AIs, the AIs will sometimes retain and use that information. Do not share information that you don’t want the AIs to use and retain, such as information about sensitive topics.
Depending on how you use AIs, deleting individual messages, entire threads, or even your account may not delete our copy of your personal information, and the Privacy Center describes the correct way to delete your personal information.
Qualifications & exceptions
Retention is stated as "sometimes", not always.
Whether deletion removes Meta’s copy depends on how you use AIs; the terms direct you to the Privacy Center for the correct deletion route.
Not stated here
How long retained information is kept, and the Privacy Center deletion instructions, are not supplied here.
Meta may review your AI conversations, automated or by humans, except chats using Private Processing
In some cases Meta will review your interactions with AIs, including the content of your conversations with or messages to AIs, and that review may be automated or manual, meaning human. The exception is Private Processing: when you interact with AI chats that use Private Processing technology, Meta says it cannot read or access the messages you have shared, including AI chats with friends and family, and that it provides Private Processing as a service on your behalf.
Supporting quotations
In some cases, Meta will review your interactions with AIs, including the content of your conversations with or messages to AIs, and this review may be automated or manual (human).
When you interact with AI chats that use Private Processing technology, Meta cannot read or access the messages you have shared, including AI chats with friends and family. Meta provides Private Processing as a service on your behalf.
Qualifications & exceptions
Review happens "in some cases"; the terms do not say which.
The Private Processing exception applies only to AI chats that use that technology.
Not stated here
Which products or chats use Private Processing is not stated here.
Sharing & Disclosure · Company practice
Using AIs instructs Meta to share information with search partners; Private Processing queries are unlinked
Meta may share certain information with select partners, such as search engines, to give you the most relevant information in connection with AIs, and the terms treat your use of AIs as an instruction to Meta to share information for that purpose. For AIs used in Private Processing, search terms based on your chat may be sent to these search partners to produce a response; the terms state those queries are not connected to your account. A linked page describes the partner sharing in more detail.
Supporting quotations
Meta may share certain information with select partners, like search engines, to provide you with the most relevant information in connection with our AIs.
By using AIs, you instruct us to share information for this purpose. To provide you with the most relevant information when using AIs in Private Processing, search terms based on your chat may be sent to these search partners to provide a response. These queries are not connected to your account.
Qualifications & exceptions
Sharing is "may", for the stated purpose of providing the most relevant information.
The not-connected-to-your-account statement is made about Private Processing search queries specifically.
Not stated here
Which partners receive information, and what "certain information" includes, is on a linked page that is not supplied here.
Cross-references
contrast (model-proposed): Meta may review your AI conversations, automated or by humans, except chats using Private Processing — contrast: Reviewed tension: Meta says it cannot read or access Private Processing messages, yet search terms based on a Private Processing chat may be sent to search partners (stated as not connected to your account). Both are stated; the terms do not explain how the search terms are derived without access.
Retention & Deletion · User right or choice
Depending on where you live, you may request access, deletion or correction of personal information
Depending on where you reside, you or an authorised agent may have the right to ask Meta for access to your personal information in a portable format, for deletion of it, or for correction of it, subject to certain exceptions. The terms point to the Privacy Center (linked, not supplied) to learn how to access and delete personal information processed in the context of AIs and to exercise any applicable rights, and say it is important to review the Privacy Center before trying to delete personal information or deleting or deactivating your account.
Supporting quotations
Depending on where you reside, you or an authorized agent may have the right to request access to your personal information from Meta in a portable format, deletion of your personal information, or correction of your personal information, subject to certain exceptions.
It is important that you review the Privacy Center before you try to delete any personal information or delete or deactivate your account.
Qualifications & exceptions
These rights depend on where you reside and are "subject to certain exceptions" that are not listed here.
Review the Privacy Center first: ordinary deletion of messages, threads or the account may not remove Meta’s copy (see the retention card).
Not stated here
Which residences carry which rights, and the exceptions, are not stated.
The Privacy Center page is linked but not supplied.
Legal Rights & Agreement Rules · Restriction or exception
No warranty of accuracy: do not rely on Outputs, and verifying them is your sole responsibility
To the maximum extent permitted by law, Meta makes no representations or warranties about the accuracy of any Content, including Outputs, nor about Outputs that are disturbing or upsetting. The terms tell you not to rely on Outputs for any purpose and not to use them to inform professional advice or decisions related to medicine, finance, law or pharmaceuticals. If you plan to use Outputs for any reason, verifying them is your sole responsibility, and the terms warn that AIs and any Content may not reflect accurate, complete or current information.
Supporting quotations
Accuracy and Warning Regarding Content
TO THE MAXIMUM EXTENT PERMITTED BY LAW, META MAKES NO REPRESENTATIONS OR WARRANTIES WITH RESPECT TO THE ACCURACY OF ANY CONTENT, INCLUDING OUTPUTS, OR ANY OUTPUTS THAT ARE DISTURBING OR UPSETTING.
YOU SHOULD NOT RELY UPON OUTPUTS FOR ANY PURPOSE OR USE OUTPUTS TO INFORM PROFESSIONAL ADVICE OR DECISIONS RELATED TO MEDICINE, FINANCE, LAW, OR PHARMACEUTICALS. IF YOU PLAN TO USE OUTPUTS FOR ANY REASON, IT IS YOUR SOLE RESPONSIBILITY TO VERIFY OUTPUTS. AIS AND ANY CONTENT MAY NOT REFLECT ACCURATE, COMPLETE, OR CURRENT INFORMATION.
Qualifications & exceptions
The disclaimer applies "to the maximum extent permitted by law", so local law may preserve some protections.
The professional-decision warning mirrors the Acceptable Uses ban on soliciting professional advice.
Legal Rights & Agreement Rules · Restriction or exception
Outputs are unpredictable and may be offensive; you, not Meta, are responsible for how you use them
AIs generate Content from your Prompts and the specific Content returned cannot be predicted in advance. The terms state, in capitals, that AIs might generate content you consider offensive or objectionable and that does not represent Meta’s views. By agreeing to the Terms and using AIs you recognise and assume the risk that your Prompts may produce such content, and you agree that you, not Meta, are responsible for your use of, your response to, and any actions you take in relation to Content generated from your Prompts.
Supporting quotations
The specific Content that the AIs will return based on your Prompts cannot be predicted in advance. AIs MIGHT GENERATE CONTENT THAT YOU CONSIDER OFFENSIVE OR OBJECTIONABLE AND/OR THAT DOES NOT REPRESENT META’S VIEWS.
YOU ALSO RECOGNIZE AND AGREE THAT YOU – NOT META – ARE RESPONSIBLE FOR YOUR USE OF, RESPONSE TO, AND/OR ANY ACTIONS YOU TAKE IN RELATION TO, CONTENT GENERATED BY THE AIS BASED ON YOUR PROMPTS.
Qualifications & exceptions
The risk you assume is framed around content generated from the Prompts you provide.
Use & Processing · Company practice
Meta uses your Content, with human review and vendors in some cases, to run and improve services and for research
Meta uses Content (your Prompts, Outputs and Feedback) and related information as described in the Meta Terms of Service and the Meta Privacy Policy. It may do so through automated or manual (human) review and, in some instances, through third-party vendors. The listed purposes include providing, maintaining and improving Meta services and features, and conducting and supporting research. This is the express carve-out from your retained rights in Prompts.
Supporting quotations
Meta’s Rights Regarding Content
Meta uses Content and related information as described in the Meta Terms of Service and Meta Privacy Policy, and may do so through automated or manual (i.e. human) review and through third-party vendors in some instances, including:
To provide, maintain, and improve Meta services and features.
To conduct and support research
Qualifications & exceptions
The list is introduced with "including", so the stated purposes are not exhaustive; the full description lives in the Meta Terms of Service and Privacy Policy.
Third-party vendors are involved "in some instances", not always.
Not stated here
The Meta Terms of Service and Meta Privacy Policy are not supplied here, so the full extent of use is unknown.
Which vendors are used, and for which purposes, is not stated.
Meta uses Content to monitor compliance, report violations required by law, remove violating Content and act on accounts
Among the listed uses of your Content, Meta may monitor your use of AIs for compliance with these Terms and applicable laws and report violations of laws or regulations where the law requires it; remove unsafe, discriminatory or other Content that violates these Terms, Meta’s Community Standards or other applicable policies and terms; and suspend or remove accounts that violate these Terms with respect to your use of AIs. Like the other uses, this may involve automated or human review and, in some instances, third-party vendors.
Supporting quotations
may do so through automated or manual (i.e. human) review and through third-party vendors in some instances, including:
To monitor your use of AIs for compliance with these Terms and applicable laws and to report violations of applicable laws or regulations as required by law
To remove unsafe, discriminatory, or other Content that violates these Terms, Meta’s Community Standards, or any other applicable policies and terms
To suspend or remove accounts that violate these Terms with respect to your use of AIs
Qualifications & exceptions
Reporting is tied to what is "required by law"; the terms do not describe voluntary reporting.
Content removal references Meta’s Community Standards and other policies that are not supplied here.
Not stated here
Meta’s Community Standards and the other applicable policies are not supplied.
Account & Service Control · Company practice
Breach may lead Meta to suspend or permanently disable AI access, other Meta services, or your account
If Meta determines that you have breached these Terms or other applicable terms and policies, it may suspend or permanently disable your access to AIs or to other Meta services or features, and it may permanently disable or delete your account, including where it is required to do so for legal reasons. The trigger is Meta’s own determination of breach; the response is discretionary and can extend beyond AIs to other Meta services.
Supporting quotations
Suspension and Termination
If we determine that you have breached these Terms or other applicable terms and policies, we may suspend or permanently disable your access to AIs or other Meta services or features, and we may permanently disable or delete your account including where we are required to do so for legal reasons.
Qualifications & exceptions
Consequences can reach other Meta services and features, not only AIs.
Account deletion includes cases where Meta is required to delete for legal reasons.
Not stated here
No notice, appeal or reinstatement procedure is described here.
Meta keeps all rights in AIs; you get only the access and use rights these Terms expressly grant
Meta retains all of its rights, title and interest, including intellectual property rights, in and to AIs. Apart from the rights of access and use that the Terms expressly grant, the Terms give you no right, title or interest in the AIs. This is the mirror of the Prompts rule: you keep the rights you have in your Prompts, and Meta keeps the rights in the AI system itself.
Supporting quotations
Ownership of AIs
We retain all our respective rights, title, and interest, including intellectual property rights, in and to AIs. Other than the rights of access and use expressly granted in these Terms, these Terms do not grant you any right, title, or interest in or to the AIs.
Except as expressly provided in these Terms, you retain all right, title, and interest – including any intellectual property rights – that you have in and to your Prompts.
Qualifications & exceptions
The only rights you receive are "the rights of access and use expressly granted in these Terms"; the Terms do not spell out a separate licence grant beyond that phrase.
Not stated here
Ownership or licensing of Outputs is not addressed in this section.
Legal Rights & Agreement Rules · Restriction or exception
Third-party content and integrations used with AIs are at your own risk; Meta is not responsible
AIs may use, or be used together with, content, services, features and integrations provided by third parties. Meta says it is not responsible for your use of that third-party content, those services, features or integrations, and that you use them at your own risk.
Supporting quotations
Third Party Materials
AIs may use or be used in connection with content, services, features, and integrations provided by third parties. Meta is not responsible for your use of third party content, services, features, and integrations, and you use these at your own risk.
Qualifications & exceptions
The statement covers both third-party material that AIs use and third-party services AIs are used in connection with.
Not stated here
Which third-party integrations are involved is not stated.
Legal Rights & Agreement Rules · Restriction or exception
AIs and Outputs come "as is": no guarantee of safety, security or accuracy; implied warranties disclaimed
AIs and Outputs are provided on an "as is" and "as available" basis. To the maximum extent permitted by applicable law, Meta makes no guarantee that they will be safe, secure or error-free or will work without disruptions, delays or imperfections; it is not responsible for Outputs, and you are responsible for checking Outputs for accuracy and suitability. Outputs do not reflect the views of Meta, its affiliates or personnel. Meta also disclaims, as far as the law allows, all express or implied warranties, listing merchantability, fitness for a particular purpose, title, accuracy, availability, security, privacy, compatibility, non-infringement and any warranty implied by course of dealing, course of performance or trade usage. The earlier accuracy warning makes the same point about Content and Outputs.
Supporting quotations
Disclaimer of Warranties and Limitations on Liability
Disclaimer of Warranties
AIs and Outputs are provided on an “as is” and “as available” basis, and, to the maximum extent permitted by applicable law, we make no guarantees that they will be safe, secure, or error-free, or will function without disruptions, delays, or imperfections. To the maximum extent permitted by law, we are not responsible for Outputs, and you are responsible for checking Outputs for accuracy and suitability. The Outputs provided by AIs do not reflect the views of Meta, our affiliates, or our personnel.
TO THE MAXIMUM EXTENT PERMITTED BY LAW, WE DISCLAIM ANY AND ALL WARRANTIES, WHETHER EXPRESS OR IMPLIED, INCLUDING THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE, ACCURACY, AVAILABILITY, SECURITY, PRIVACY, COMPATIBILITY, NON-INFRINGEMENT, AND ANY WARRANTY IMPLIED BY COURSE OF DEALING, COURSE OF PERFORMANCE, OR TRADE USAGE.
TO THE MAXIMUM EXTENT PERMITTED BY LAW, META MAKES NO REPRESENTATIONS OR WARRANTIES WITH RESPECT TO THE ACCURACY OF ANY CONTENT, INCLUDING OUTPUTS, OR ANY OUTPUTS THAT ARE DISTURBING OR UPSETTING.
Qualifications & exceptions
Each disclaimer is limited "to the maximum extent permitted by (applicable) law", so some protections may survive under local law.
Responsibility for checking Outputs for accuracy and suitability is placed on you.
Legal Rights & Agreement Rules · Restriction or exception
Meta excludes lost profits, data and indirect damages, with a fallback where local law forbids such limits
Meta’s liability is limited to the fullest extent permitted by applicable law. Under no circumstance will Meta be liable to you for lost profits, revenues, information or data, or for consequential, special, indirect, exemplary, punitive or incidental damages arising out of or related to these Terms or AIs, however caused and on any theory of liability including negligence, even if Meta was advised such damages were possible. The terms acknowledge that the laws of some states or jurisdictions may not allow some of these exclusions, so some or all may not apply to you; in that case Meta’s liability is still limited to the fullest extent the applicable law permits.
Supporting quotations
Limitations on Liability
Our liability shall be limited to the fullest extent permitted by applicable law, and under no circumstance will we be liable to you for any lost profits, revenues, information, or data, or consequential, special, indirect, exemplary, punitive, or incidental damages arising out of or related to these Terms or AIs (however caused and on any theory of liability, including negligence), even if we have been advised of the possibility of such damages.
The laws of some states or jurisdictions may not allow the exclusion or limitation of certain damages, so some or all of the exclusions and limitations set out above may not apply to you. Notwithstanding anything to the contrary in these Terms, in such cases, Meta’s liability will be limited to the fullest extent permitted by applicable law.
Qualifications & exceptions
Where local law does not allow a given exclusion or limitation, that part may not apply to you, and the fallback is the fullest limitation the law does permit.
No monetary cap figure is stated.
Not stated here
Which jurisdictions’ laws override the exclusions is not stated.
Legal Rights & Agreement Rules · Procedure
Disputes under these Terms follow the dispute resolution clause of the Meta Terms of Service
Any claim, cause of action or dispute that arises out of or relates to these Terms is subject to the dispute resolution clause in the Meta Terms of Service. That clause is not reproduced here; the Meta Terms of Service are a separate document which, under the scope section, you also agree to by using AIs.
Supporting quotations
Disputes
Any claim, cause of action, or dispute that arises out of or relates to these Terms is subject to the disputes resolution clause in the Meta Terms of Service.
By using AIs, you agree to be bound by these Terms, the
Qualifications & exceptions
The governing procedure (for example arbitration, courts or venue) is whatever the Meta Terms of Service dispute clause provides.
Not stated here
The Meta Terms of Service dispute resolution clause is not supplied here; its content, including any arbitration requirement or opt-out, is unknown from this source.
Unenforceable parts are trimmed or severed; non-enforcement is not waiver; changes need Meta’s signed writing
If any portion of these Terms is found unenforceable, it is deemed amended to the minimum extent necessary to make it enforceable, and if that is not possible it is severed while the rest stays in full force. Meta’s failure to enforce any term is not a waiver of it. Any amendment to or waiver of these Terms must be in writing and signed by Meta.
Supporting quotations
Severability
If any portion of these Terms is found to be unenforceable, the unenforceable portion will be deemed amended to the minimum extent necessary to make it enforceable, and if it can’t be made enforceable, then it will be severed and the remaining portion will remain in full force and effect. If we fail to enforce any of these Terms, it will not be considered a waiver. Any amendment to or waiver of these Terms must be made in writing and signed by us.
Qualifications & exceptions
The signed-writing requirement is for amendments and waivers of these Terms; the Changes section describes how Meta itself issues updated Terms.
Legal Rights & Agreement Rules · Company commitment
Meta may update these Terms; material changes come with notice and a review period unless law requires otherwise
Meta says it may need to update these Terms from time to time to reflect its services and practices, to promote a safe and secure experience, or to comply with applicable law. Unless the law requires otherwise, Meta will notify you of any material changes, for example by email or in-app notification, and give you an opportunity to review them before they take effect. Once updated Terms are in effect you are bound by them if you continue to use AIs; if you do not agree, you should stop using AIs before they take effect.
Supporting quotations
Changes to These Terms
we may need to update these Terms from time to time to accurately reflect our services and practices, to promote a safe and secure experience, and/or to comply with applicable law
Unless otherwise required by law, we will notify you (for example through email or in-app notification) of any material changes to these Terms and give you an opportunity to review them before they go into effect. Once any updated Terms are in effect you will be bound by them if you continue to use our AIs. If you do not agree to the updated Terms, you should stop using AIs before they go into effect.
Qualifications & exceptions
Notice and a review opportunity are promised for "material" changes; non-material changes are not covered by the promise.
"Unless otherwise required by law" is the stated exception to advance notice.
Continued use after the effective date is treated as acceptance; the remedy for disagreement is to stop using AIs before then.
Not stated here
What counts as a material change, and how long the review period is, are not stated.