Defense in depth across input, output, behaviour, labelers, and reports.
ourdream’s content moderation is built as
defense in depth:
multiple independent layers stacked together, each tuned to catch a
different category of content. Each layer specialises in a different
class of signal, and the layers reinforce each other.
The layers
1. Input
Every prompt is classified before any model runs against it. Chat
messages, image prompts, and character submissions pass through a
set of classifiers covering content, behaviour, and metadata signals.
Inputs that match the strictest signals are blocked before the model
runs at all; other flagged inputs are routed to the next layer.
2. Output
Generated images go through a separate set of classifiers that looks
at the output independent of the prompt. This is an independent check
on what the model actually produces, covering the categories that map
to Prohibited content. Outputs that
fail are blocked from delivery and may trigger account level review.
3. Metadata and behaviour
The third layer reads patterns rather than content. It looks at signals
across many actions by the same user, or across many users producing
similar content. Examples include accounts that produce content the
other layers reject at unusually high rates, and prompts that share
structural features with previously-blocked ones.Behaviour signals usually do not block content on their own. They raise
the priority of human review.
4. Human reviewers
Trained reviewers see what the classifiers flag. Reviewers can:
Approve content (release it from review).
Reject it with a reason that maps to a specific policy.
Escalate ambiguous cases to a senior reviewer or to the trust team.
Flag patterns that the classifier layers should learn from; edge
cases feed back into the rules and training data.
For public character submissions the review queue runs in the
hundreds per day. For chat-generated content the volume is much
higher; the classifier layers handle the bulk of it, and the cases
that warrant a human reach a human.
5. Community reports
The last layer is everyone using the product. Users can
report characters, images, videos, and
scenarios. Reports trigger re-review against the current policy,
which may be stricter than the one in force when the content was first
generated.The most severe categories (underage content) are reviewed first.
Why this shape
Each layer specialises. Classifiers handle volume at speed. Metadata
signals catch behavioural patterns. Human reviewers handle ambiguity
and judgement. The community surfaces what’s actually playing out in
practice. Together they cover more ground than any single layer could
alone.
The moderation team
Reviewers are vetted ourdream staff and trusted long-term community
contributors who have been onboarded into the moderation rota. They
operate under written guidelines and undergo regular calibration with
each other and with the trust team to keep judgement consistent.Moderation decisions are made independent of growth and revenue
metrics. The trust team owns the policy; product owns the product.We care about moderator wellbeing because the work is hard, and do
not tolerate abuse or harassment of our moderation team. The team
also meets regularly to review edge cases and update the guidelines.
What gets rejected most often
In the public-submission queue, in rough order of frequency:
Profile images with underage cues.
Characters too close to existing IP.
Scenarios that frame non-consent as desirable — whether the user
is cast as perpetrator or the framing otherwise presents
non-consent as the appeal.
Names or descriptions matching the real-person blocklist.
Automated classifiers process all content on the platform —
including private chat messages and private generations — because
they have to in order to enforce the policies above. Classifier
processing happens in-line and no human reads the content as part of
it.Human reviewers only see content that has been (a) flagged by a
classifier, (b) reported by a user, or (c) escalated as part of a
trust-team investigation. They do not browse private chats. Account
data (email, payment information, IP address) is gated behind trust-
team access and is touched only when an investigation requires it.
Appeals
If a creator believes a rejection or removal was applied incorrectly,
the Appeals process is the way to raise it. A
different reviewer handles the appeal, with the original objection in
mind. Appeals also feed back into the rules and training data, so the
work creators put into making the case rarely goes nowhere.
Last modified on June 9, 2026
{"docsConfig":{"theme":"mint","$schema":"https://mintlify.com/docs.json","name":"ourdream Trust \u0026 Safety","description":"Our approach to safety, moderation, and trust at ourdream.","colors":{"primary":"#FD5FC2","light":"#F17BB6","dark":"#FD5FC2"},"logo":{"light":"https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/logo/light.svg?fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=01fa9593e7f718b95a67ca241d25508b","dark":"https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/logo/dark.svg?fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=93b24efa461eeb506f59d30125e3be98","href":"https://ourdream.ai"},"favicon":"https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/favicon.svg?fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=151bc91283ad3aebc86464a86b72f4a8","appearance":{"default":"dark","strict":false},"background":{"color":{"light":"#FFFFFF","dark":"#0D0D0D"}},"navbar":{"links":[{"href":"https://ourdream.ai","label":"ourdream"}],"primary":{"type":"button","label":"Report a problem","href":"/reporting/how-to-report"}},"footer":{"socials":{"website":"https://ourdream.ai"},"links":[{"header":"Contact","items":[{"label":"trust@ourdream.ai","href":"mailto:trust@ourdream.ai"},{"label":"ourdream.ai","href":"https://ourdream.ai"}]}]},"fonts":{"family":"Inter"},"metadata":{"timestamp":true},"styling":{"latex":false},"navigation":{"pages":[]}},"deployment":{"deploymentMetadataValue":{"subdomain":"od-f1eb486d","hasAudioNarration":false,"actualSubdomain":"od-f1eb486d","customDomains":["safety.ourdream.ai"],"gitSource":{"type":"github","owner":"ourdreamai","repo":"ourdream","deployBranch":"main","contentDirectory":"apps/safety-site","isPrivate":true,"mountPath":"","label":""},"entitlements":{"AI_CHAT":{"status":"DISABLED"}},"buildId":"6ad1d18cb4857fd9ffe6ca83","clientVersion":"0.0.4205","searchFilterCounts":[{"language":"default","tag":"Trust \u0026 Safety","count":14},{"language":"default","tag":"Overview","count":3},{"language":"default","tag":"Moderation","count":3},{"language":"default","tag":"Policies","count":3},{"language":"default","tag":"Your account","count":3},{"language":"default","tag":"Contact","count":1},{"language":"default","tag":"Reporting","count":1}]},"assistantConfig":{"enableAskAiButton":false,"enableStarterQuestions":false,"enableContextualStarterQuestions":false,"enableAskAiSearchBar":true},"basePath":"","hasChatPermissions":false,"hasWebMcpPermissions":false,"toggles":[],"googleFontsInlined":false},"customCss":[],"page":{"pagePath":"moderation/how-it-works","slug":"/moderation/how-it-works","pageMetadata":{"title":"How moderation works","description":"Defense in depth across input, output, behaviour, labelers, and reports.","href":"/moderation/how-it-works"},"description":{"compiledSource":"\"use strict\";\nconst {jsx: _jsx} = arguments[0];\nconst {useMDXComponents: _provideComponents} = arguments[0];\nfunction _createMdxContent(props) {\n const _components = {\n p: \"p\",\n ..._provideComponents(),\n ...props.components\n };\n return _jsx(_components.p, {\n children: \"Defense in depth across input, output, behaviour, labelers, and reports.\"\n });\n}\nfunction MDXContent(props = {}) {\n const {wrapper: MDXLayout} = {\n ..._provideComponents(),\n ...props.components\n };\n return MDXLayout ? _jsx(MDXLayout, {\n ...props,\n children: _jsx(_createMdxContent, {\n ...props\n })\n }) : _createMdxContent(props);\n}\nreturn {\n default: MDXContent\n};\n","frontmatter":{},"scope":{}},"pageType":"default","mdx":{"mdxSource":{"compiledSource":"\"use strict\";const{Fragment:_Fragment,jsx:_jsx,jsxs:_jsxs}=arguments[0];const{useMDXComponents:_provideComponents}=arguments[0];function _createMdxContent(props){const _components={a:`a`,em:`em`,li:`li`,ol:`ol`,p:`p`,strong:`strong`,ul:`ul`,..._provideComponents(),...props.components},{Heading,OptimizedImage}=_components;if(!Heading)_missingMdxReference(`Heading`,true);if(!OptimizedImage)_missingMdxReference(`OptimizedImage`,true);return _jsxs(_Fragment,{children:[_jsxs(_components.p,{children:[`ourdream’s content moderation is built as\n`,_jsx(_components.a,{href:`https://en.wikipedia.org/wiki/Defense_in_depth_(computing)`,children:_jsx(_components.strong,{children:`defense in depth`})}),`:\nmultiple independent layers stacked together, each tuned to catch a\ndifferent category of content. Each layer specialises in a different\nclass of signal, and the layers reinforce each other.`]}),`\n`,_jsx(OptimizedImage,{className:`mint-block dark:mint-hidden`,src:`https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-light.svg?fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=1e33ffabf9ab5f8c7a0a8bc5388ef698`,alt:`Defense in depth: five moderation layers stacked. Input, output, metadata and behaviour, human reviewers, community reports.`,\"data-og-width\":`720`,width:`720`,\"data-og-height\":`460`,height:`460`,\"data-path\":`images/defense-in-depth-light.svg`,\"data-optimize\":`true`,\"data-opv\":`3`,srcset:`https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-light.svg?w=280\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=ede7f546922c913a560287770a89974f 280w, https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-light.svg?w=560\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=44fec15b7d3367d45541a050dd06a14f 560w, https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-light.svg?w=840\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=d054a9cbb7cd297ebadab7cc502aa2a4 840w, https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-light.svg?w=1100\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=ec070df9811963a24a9fb1bdb616d329 1100w, https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-light.svg?w=1650\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=ce32268b8b55b06a9784fba83190051c 1650w, https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-light.svg?w=2500\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=c97967a7b2202a54a3f79cf152f06fc4 2500w`}),`\n`,_jsx(OptimizedImage,{className:`mint-hidden dark:mint-block`,src:`https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-dark.svg?fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=6946f2164ceb77b0d3bb425442d83d28`,alt:`Defense in depth: five moderation layers stacked. Input, output, metadata and behaviour, human reviewers, community reports.`,\"data-og-width\":`720`,width:`720`,\"data-og-height\":`460`,height:`460`,\"data-path\":`images/defense-in-depth-dark.svg`,\"data-optimize\":`true`,\"data-opv\":`3`,srcset:`https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-dark.svg?w=280\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=a237d40dad925d62701dc0fdf3baacf8 280w, https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-dark.svg?w=560\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=c2624e0c5dbf560a0ca2cbee6d7f2f3e 560w, https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-dark.svg?w=840\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=67c50cdd58d135e1ecd1a3d630527e3c 840w, https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-dark.svg?w=1100\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=45dbb9ea13ac77781422fcee3aa3ccec 1100w, https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-dark.svg?w=1650\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=e67e477e5a040170659929e092e16093 1650w, https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/images/defense-in-depth-dark.svg?w=2500\u0026fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=370e370539879167cdb82bae0cceccb1 2500w`}),`\n`,_jsx(Heading,{level:`2`,id:`the-layers`,children:`The layers`}),`\n`,_jsx(Heading,{level:`3`,id:`1-input`,children:`1. Input`}),`\n`,_jsx(_components.p,{children:`Every prompt is classified before any model runs against it. Chat\nmessages, image prompts, and character submissions pass through a\nset of classifiers covering content, behaviour, and metadata signals.\nInputs that match the strictest signals are blocked before the model\nruns at all; other flagged inputs are routed to the next layer.`}),`\n`,_jsx(Heading,{level:`3`,id:`2-output`,children:`2. Output`}),`\n`,_jsxs(_components.p,{children:[`Generated images go through a separate set of classifiers that looks\nat the output independent of the prompt. This is an independent check\non what the model actually produces, covering the categories that map\nto `,_jsx(_components.a,{href:`/policies/prohibited-content`,children:`Prohibited content`}),`. Outputs that\nfail are blocked from delivery and may trigger account level review.`]}),`\n`,_jsx(Heading,{level:`3`,id:`3-metadata-and-behaviour`,children:`3. Metadata and behaviour`}),`\n`,_jsx(_components.p,{children:`The third layer reads patterns rather than content. It looks at signals\nacross many actions by the same user, or across many users producing\nsimilar content. Examples include accounts that produce content the\nother layers reject at unusually high rates, and prompts that share\nstructural features with previously-blocked ones.`}),`\n`,_jsx(_components.p,{children:`Behaviour signals usually do not block content on their own. They raise\nthe priority of human review.`}),`\n`,_jsx(Heading,{level:`3`,id:`4-human-reviewers`,children:`4. Human reviewers`}),`\n`,_jsx(_components.p,{children:`Trained reviewers see what the classifiers flag. Reviewers can:`}),`\n`,_jsxs(_components.ul,{children:[`\n`,_jsx(_components.li,{children:`Approve content (release it from review).`}),`\n`,_jsx(_components.li,{children:`Reject it with a reason that maps to a specific policy.`}),`\n`,_jsx(_components.li,{children:`Escalate ambiguous cases to a senior reviewer or to the trust team.`}),`\n`,_jsx(_components.li,{children:`Flag patterns that the classifier layers should learn from; edge\ncases feed back into the rules and training data.`}),`\n`]}),`\n`,_jsx(_components.p,{children:`For public character submissions the review queue runs in the\nhundreds per day. For chat-generated content the volume is much\nhigher; the classifier layers handle the bulk of it, and the cases\nthat warrant a human reach a human.`}),`\n`,_jsx(Heading,{level:`3`,id:`5-community-reports`,children:`5. Community reports`}),`\n`,_jsxs(_components.p,{children:[`The last layer is everyone using the product. Users can\n`,_jsx(_components.a,{href:`/reporting/how-to-report`,children:`report`}),` characters, images, videos, and\nscenarios. Reports trigger re-review against the `,_jsx(_components.em,{children:`current`}),` policy,\nwhich may be stricter than the one in force when the content was first\ngenerated.`]}),`\n`,_jsx(_components.p,{children:`The most severe categories (underage content) are reviewed first.`}),`\n`,_jsx(Heading,{level:`2`,id:`why-this-shape`,children:`Why this shape`}),`\n`,_jsx(_components.p,{children:`Each layer specialises. Classifiers handle volume at speed. Metadata\nsignals catch behavioural patterns. Human reviewers handle ambiguity\nand judgement. The community surfaces what’s actually playing out in\npractice. Together they cover more ground than any single layer could\nalone.`}),`\n`,_jsx(Heading,{level:`2`,id:`the-moderation-team`,children:`The moderation team`}),`\n`,_jsx(_components.p,{children:`Reviewers are vetted ourdream staff and trusted long-term community\ncontributors who have been onboarded into the moderation rota. They\noperate under written guidelines and undergo regular calibration with\neach other and with the trust team to keep judgement consistent.`}),`\n`,_jsx(_components.p,{children:`Moderation decisions are made independent of growth and revenue\nmetrics. The trust team owns the policy; product owns the product.`}),`\n`,_jsx(_components.p,{children:`We care about moderator wellbeing because the work is hard, and do\nnot tolerate abuse or harassment of our moderation team. The team\nalso meets regularly to review edge cases and update the guidelines.`}),`\n`,_jsx(Heading,{level:`2`,id:`what-gets-rejected-most-often`,children:`What gets rejected most often`}),`\n`,_jsx(_components.p,{children:`In the public-submission queue, in rough order of frequency:`}),`\n`,_jsxs(_components.ol,{children:[`\n`,_jsx(_components.li,{children:`Profile images with underage cues.`}),`\n`,_jsx(_components.li,{children:`Characters too close to existing IP.`}),`\n`,_jsx(_components.li,{children:`Scenarios that frame non-consent as desirable — whether the user\nis cast as perpetrator or the framing otherwise presents\nnon-consent as the appeal.`}),`\n`,_jsx(_components.li,{children:`Names or descriptions matching the real-person blocklist.`}),`\n`]}),`\n`,_jsxs(_components.p,{children:[`Each of these is described on\n`,_jsx(_components.a,{href:`/policies/prohibited-content`,children:`Prohibited content`}),`.`]}),`\n`,_jsx(Heading,{level:`2`,id:`what-classifiers-see-vs-what-humans-see`,children:`What classifiers see vs what humans see`}),`\n`,_jsxs(_components.p,{children:[`Automated classifiers process `,_jsx(_components.strong,{children:`all`}),` content on the platform —\nincluding private chat messages and private generations — because\nthey have to in order to enforce the policies above. Classifier\nprocessing happens in-line and no human reads the content as part of\nit.`]}),`\n`,_jsxs(_components.p,{children:[_jsx(_components.strong,{children:`Human reviewers`}),` only see content that has been (a) flagged by a\nclassifier, (b) reported by a user, or (c) escalated as part of a\ntrust-team investigation. They do not browse private chats. Account\ndata (email, payment information, IP address) is gated behind trust-\nteam access and is touched only when an investigation requires it.`]}),`\n`,_jsx(Heading,{level:`2`,id:`appeals`,children:`Appeals`}),`\n`,_jsxs(_components.p,{children:[`If a creator believes a rejection or removal was applied incorrectly,\nthe `,_jsx(_components.a,{href:`/moderation/appeals`,children:`Appeals`}),` process is the way to raise it. A\ndifferent reviewer handles the appeal, with the original objection in\nmind. Appeals also feed back into the rules and training data, so the\nwork creators put into making the case rarely goes nowhere.`]})]})}function MDXContent(props={}){const{wrapper:MDXLayout}={..._provideComponents(),...props.components};return MDXLayout?_jsx(MDXLayout,{...props,children:_jsx(_createMdxContent,{...props})}):_createMdxContent(props)}return{default:MDXContent};function _missingMdxReference(id,component){throw new Error(`Expected `+(component?`component`:`object`)+\" `\"+id+\"` to be defined: you likely forgot to import, pass, or provide it.\")}","frontmatter":{},"scope":{"config":{},"pageMetadata":{"title":"How moderation works","description":"Defense in depth across input, output, behaviour, labelers, and reports.","href":"/moderation/how-it-works"}}},"mdxSourceWithNoJs":null},"mdxExtracts":{"tableOfContents":[{"title":"The layers","slug":"the-layers","depth":2,"children":[{"title":"1. Input","slug":"1-input","depth":3,"children":[]},{"title":"2. Output","slug":"2-output","depth":3,"children":[]},{"title":"3. Metadata and behaviour","slug":"3-metadata-and-behaviour","depth":3,"children":[]},{"title":"4. Human reviewers","slug":"4-human-reviewers","depth":3,"children":[]},{"title":"5. Community reports","slug":"5-community-reports","depth":3,"children":[]}]},{"title":"Why this shape","slug":"why-this-shape","depth":2,"children":[]},{"title":"The moderation team","slug":"the-moderation-team","depth":2,"children":[]},{"title":"What gets rejected most often","slug":"what-gets-rejected-most-often","depth":2,"children":[]},{"title":"What classifiers see vs what humans see","slug":"what-classifiers-see-vs-what-humans-see","depth":2,"children":[]},{"title":"Appeals","slug":"appeals","depth":2,"children":[]}],"codeExamples":{}},"jsLinks":[],"dynamicCss":".mint-block {\n display: block\n}\n.mint-hidden {\n display: none\n}\n.dark\\:mint-block:is(.dark *) {\n display: block\n}\n.dark\\:mint-hidden:is(.dark *) {\n display: none\n}","subdomain":"od-f1eb486d","siteName":"ourdream Trust \u0026 Safety","title":"How moderation works - ourdream Trust \u0026 Safety","navMaps":{"firstHrefInVersion":{},"firstHrefInLanguage":{}},"lastModified":"2026-06-10T04:43:43.217Z","gitLastModified":"2026-06-09T00:45:55.000Z","originalFileLocation":"moderation/how-it-works.mdx"},"headHtml":"\u003cscript data-mint-head>(function(){var d=document.documentElement,p=null;if(!p){try{p=localStorage.getItem(\"isDarkMode\")}catch(e){}if(p==='true')p='dark';else if(p==='false')p='light';if(p!=='light'\u0026\u0026p!=='dark'\u0026\u0026p!=='system')p=\"dark\"}var r=p==='system'?(matchMedia(\"(prefers-color-scheme: dark)\").matches?'dark':'light'):p;d.classList.remove(r==='dark'?'light':'dark');d.classList.add(r);d.setAttribute(\"data-theme-preference\",p)})();\u003c/script>\u003cmeta data-mint-head name=\"description\" content=\"Defense in depth across input, output, behaviour, labelers, and reports.\">\u003cmeta data-mint-head name=\"application-name\" content=\"ourdream Trust \u0026amp; Safety\">\u003cmeta data-mint-head name=\"generator\" content=\"Mintlify\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"192x192\" href=\"/mintlify-assets/_mintlify/favicons/od-f1eb486d/4mQv6U33Ut5YC6I5/_generated/favicon/android-chrome-192x192.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"apple-touch-icon\" type=\"image/png\" sizes=\"180x180\" href=\"/mintlify-assets/_mintlify/favicons/od-f1eb486d/4mQv6U33Ut5YC6I5/_generated/favicon/apple-touch-icon.png\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"16x16\" href=\"/mintlify-assets/_mintlify/favicons/od-f1eb486d/4mQv6U33Ut5YC6I5/_generated/favicon/favicon-16x16.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"32x32\" href=\"/mintlify-assets/_mintlify/favicons/od-f1eb486d/4mQv6U33Ut5YC6I5/_generated/favicon/favicon-32x32.png\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"shortcut icon\" type=\"image/x-icon\" href=\"/mintlify-assets/_mintlify/favicons/od-f1eb486d/4mQv6U33Ut5YC6I5/_generated/favicon/favicon.ico\" media=\"(prefers-color-scheme: light)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"192x192\" href=\"/mintlify-assets/_mintlify/favicons/od-f1eb486d/4mQv6U33Ut5YC6I5/_generated/favicon-dark/android-chrome-192x192.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"16x16\" href=\"/mintlify-assets/_mintlify/favicons/od-f1eb486d/4mQv6U33Ut5YC6I5/_generated/favicon-dark/favicon-16x16.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"icon\" type=\"image/png\" sizes=\"32x32\" href=\"/mintlify-assets/_mintlify/favicons/od-f1eb486d/4mQv6U33Ut5YC6I5/_generated/favicon-dark/favicon-32x32.png\" media=\"(prefers-color-scheme: dark)\">\u003clink data-mint-head rel=\"shortcut icon\" type=\"image/x-icon\" href=\"/mintlify-assets/_mintlify/favicons/od-f1eb486d/4mQv6U33Ut5YC6I5/_generated/favicon-dark/favicon.ico\" media=\"(prefers-color-scheme: dark)\">\u003cmeta data-mint-head property=\"og:title\" content=\"How moderation works - ourdream Trust \u0026amp; Safety\">\u003cmeta data-mint-head property=\"og:description\" content=\"Defense in depth across input, output, behaviour, labelers, and reports.\">\u003cmeta data-mint-head property=\"og:site_name\" content=\"ourdream Trust \u0026amp; Safety\">\u003cmeta data-mint-head property=\"og:url\" content=\"https://safety.ourdream.ai/moderation/how-it-works\">\u003cmeta data-mint-head property=\"og:type\" content=\"website\">\u003cmeta data-mint-head property=\"og:image\" content=\"https://od-f1eb486d.mintlify.site/_mintlify/api/og/v2?division=Moderation\u0026amp;title=How+moderation+works\u0026amp;description=Defense+in+depth+across+input%2C+output%2C+behaviour%2C+labelers%2C+and+reports.\u0026amp;theme=671900b1410780aab6629e6c\u0026amp;signature=343f6869d03f3b45e9a10992a8ddeb7b391aada5610abda6afbb6426a734c6ed\">\u003cmeta data-mint-head property=\"og:image:width\" content=\"1200\">\u003cmeta data-mint-head property=\"og:image:height\" content=\"630\">\u003cmeta data-mint-head name=\"twitter:card\" content=\"summary_large_image\">\u003cmeta data-mint-head name=\"twitter:title\" content=\"How moderation works - ourdream Trust \u0026amp; Safety\">\u003cmeta data-mint-head name=\"twitter:description\" content=\"Defense in depth across input, output, behaviour, labelers, and reports.\">\u003cmeta data-mint-head name=\"twitter:image\" content=\"https://od-f1eb486d.mintlify.site/_mintlify/api/og/v2?division=Moderation\u0026amp;title=How+moderation+works\u0026amp;description=Defense+in+depth+across+input%2C+output%2C+behaviour%2C+labelers%2C+and+reports.\u0026amp;theme=671900b1410780aab6629e6c\u0026amp;signature=343f6869d03f3b45e9a10992a8ddeb7b391aada5610abda6afbb6426a734c6ed\">\u003cmeta data-mint-head name=\"twitter:image:width\" content=\"1200\">\u003cmeta data-mint-head name=\"twitter:image:height\" content=\"630\">\u003cmeta data-mint-head name=\"msapplication-config\" content=\"/mintlify-assets/_mintlify/favicons/od-f1eb486d/4mQv6U33Ut5YC6I5/_generated/favicon/browserconfig.xml\">\u003cmeta data-mint-head name=\"apple-mobile-web-app-title\" content=\"ourdream Trust \u0026amp; Safety\">\u003cmeta data-mint-head name=\"msapplication-TileColor\" content=\"#FD5FC2\">\u003clink data-mint-head rel=\"canonical\" href=\"https://safety.ourdream.ai/moderation/how-it-works\">\u003clink data-mint-head rel=\"alternate\" type=\"application/xml\" href=\"/sitemap.xml\">\u003clink data-mint-head rel=\"alternate\" type=\"text/markdown\" href=\"/moderation/how-it-works.md\">\u003cscript data-mint-head type=\"application/ld+json\">{\"@context\":\"https://schema.org\",\"@graph\":[{\"@type\":\"Organization\",\"@id\":\"https://safety.ourdream.ai/#organization\",\"name\":\"ourdream Trust \u0026 Safety\",\"url\":\"https://safety.ourdream.ai\",\"logo\":{\"@type\":\"ImageObject\",\"url\":\"https://mintcdn.com/od-f1eb486d/_8zgpEaEHP3ySGuf/logo/light.svg?fit=max\u0026auto=format\u0026n=_8zgpEaEHP3ySGuf\u0026q=85\u0026s=01fa9593e7f718b95a67ca241d25508b\"}},{\"@type\":\"WebSite\",\"@id\":\"https://safety.ourdream.ai#website\",\"name\":\"ourdream Trust \u0026 Safety\",\"url\":\"https://safety.ourdream.ai\",\"publisher\":{\"@id\":\"https://safety.ourdream.ai/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https://safety.ourdream.ai/moderation/how-it-works#webpage\",\"url\":\"https://safety.ourdream.ai/moderation/how-it-works\",\"name\":\"How moderation works\",\"description\":\"Defense in depth across input, output, behaviour, labelers, and reports.\",\"dateModified\":\"2026-06-09T00:45:55.000Z\",\"isPartOf\":{\"@id\":\"https://safety.ourdream.ai#website\"},\"breadcrumb\":{\"@id\":\"https://safety.ourdream.ai/moderation/how-it-works#breadcrumb\"}},{\"@type\":\"BreadcrumbList\",\"@id\":\"https://safety.ourdream.ai/moderation/how-it-works#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"ourdream Trust \u0026 Safety\",\"item\":\"https://safety.ourdream.ai\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"How moderation works\",\"item\":\"https://safety.ourdream.ai/moderation/how-it-works\"}]},{\"@type\":[\"Article\",\"TechArticle\"],\"@id\":\"https://safety.ourdream.ai/moderation/how-it-works#article\",\"headline\":\"How moderation works\",\"name\":\"How moderation works\",\"description\":\"Defense in depth across input, output, behaviour, labelers, and reports.\",\"url\":\"https://safety.ourdream.ai/moderation/how-it-works\",\"mainEntityOfPage\":{\"@id\":\"https://safety.ourdream.ai/moderation/how-it-works#webpage\"},\"dateModified\":\"2026-06-09T00:45:55.000Z\",\"publisher\":{\"@id\":\"https://safety.ourdream.ai/#organization\"},\"isPartOf\":{\"@id\":\"https://safety.ourdream.ai#website\"}}]}\u003c/script>\u003cscript data-mint-head type=\"application/ld+json\">{\"@context\":\"https://schema.org\",\"@type\":\"WebSite\",\"name\":\"ourdream Trust \u0026 Safety\",\"publisher\":{\"@type\":\"Organization\",\"name\":\"ourdream Trust \u0026 Safety\"},\"provider\":{\"@type\":\"Organization\",\"name\":\"Mintlify\",\"url\":\"https://mintlify.com\"}}\u003c/script>","agentPointerHref":"/moderation/how-it-works.md","siteName":"ourdream Trust \u0026 Safety","nav":{"global":null,"tabs":[{"tab":"Trust \u0026 Safety","groups":[{"group":"Overview","pages":[{"title":"Our approach to safety at ourdream","href":"/introduction"},{"title":"Principles","href":"/principles"},{"title":"What we won't do","href":"/policies/what-we-wont-do"}]},{"group":"Policies","pages":[{"title":"Prohibited content","href":"/policies/prohibited-content"},{"title":"Age verification","href":"/policies/age-verification"},{"title":"Intellectual property","description":"How ourdream handles named characters, real people, deepfakes, and copyrighted worlds.","href":"/policies/intellectual-property"}]},{"group":"Moderation","pages":[{"title":"How moderation works","href":"/moderation/how-it-works"},{"title":"Why was my character rejected?","description":"The most common rejection reasons, with examples and the fix for each.","href":"/moderation/why-rejected"},{"title":"Appeals","href":"/moderation/appeals"}]},{"group":"Reporting","pages":[{"title":"Report a problem","href":"/reporting/how-to-report"}]},{"group":"Your account","pages":[{"title":"Your safety tools","href":"/your-account/safety-tools"},{"title":"Wellbeing resources","href":"/your-account/wellbeing-resources"},{"title":"Privacy at a glance","href":"/your-account/privacy-summary"}]},{"group":"Contact","pages":[{"title":"Contact","href":"/contact"}]}]}]},"basePath":"","title":"How moderation works - ourdream Trust \u0026 Safety","deploymentId":"dpl_FZMKUpaaKiVETbfqhoUe4sPFDTmY","buildId":"dpl_FZMKUpaaKiVETbfqhoUe4sPFDTmY","dynamics":["a74be98ba7"],"chunks":["zoom-image-lightbox-B9O1U98p.js","lightbox-handle-r17bVUz3.js"]}