18+ · candid · private
Kindroid NSFW: what the Kindroid app allows, what it filters, pricing, and a private alternative
Search for whether Kindroid allows NSFW and you get a wall of affiliate roundups, none of which appear to have opened the product documentation. That is partly Kindroid fault: its help center is a JavaScript application, so the pages look empty to most crawlers and to anyone reading the raw HTML. The content is genuinely there, and it is unusually detailed.
Who do you want to talk to?
You can change who you talk to anytime.
18+ · Private and anonymous · No minors, non-consent, or anything illegal.
102+ people already talking freely
Short answer
Kindroid allows NSFW and erotic roleplay in text, and says so in writing: its moderation guidelines list "NSFW (Not Safe for Work) and Erotic Roleplay (ERP)" under a heading called "What is Explicitly Allowed." There is no NSFW toggle for chat, because there is no chat filter to switch. The toggle people are looking for does exist, but only for media and only in one place: on the website you can pick an SFW or NSFW engine for video selfies, and Kindroid states the phone apps run a keyword-based NSFW filter that the web version does not have, as "a precaution for certain app store policies dictated by Apple and Google." Moderation targets accounts rather than messages, against three named red lines, with a warning before any lock. Standard access runs $13.99 a month on the web and $15.99 through the app stores.
Last updated September 2026
On September 2, 2026 we read eleven of those help pages end to end: the moderation guidelines, the age verification policy, the subscription and in-app purchase pricing, the selfie and video documentation, and the FAQ. What follows is what Kindroid itself says, quoted, with the parts that surprised us called out. The short version is that the question people keep typing into search engines, where is the Kindroid NSFW toggle, has a real answer, and it is not the one the roundups give.
What makes it feel honest
Kindroid NSFW, the way it should feel
The filter is on the app, not the AI
Kindroid documents a keyword filter on media generation that exists in its phone apps and not on its website. Its own reason: app store policy. That pattern repeats across this whole market.
Published limits beat guessed ones
Kindroid names three prohibited categories and works through examples of each. We name three hard lines and print them on every page. Both beat a vendor that will not say.
Browser only, by choice
We never shipped a phone app, so no app store reviewer sets our content rules. That is the same reason Kindroid web behaves differently from Kindroid mobile.
Honest comparison
Where each platform actually puts the limit: on the model, on the account, or on the app build
| Platform | Is the text model itself filtered? | Where the media limit sits | What actually gets an account actioned | Primary source we read |
|---|---|---|---|---|
| Kindroid | No. Its FAQ calls it "a neutrally aligned, unfiltered AI" and the moderation guidelines say "we are not implementing a filter on the AI itself." | On the app build. A "keyword-based NSFW filter" runs in the iOS and Android apps "but not when Kindroid is accessed from the website," and the SFW or NSFW video engine choice is offered "on web only." | Three named red lines: imminent self-harm, imminent harm to others, and CSAM. Warning first, then a lock. NSFW and erotic roleplay are listed as explicitly allowed. | kindroid.ai help center, eleven articles, read September 2, 2026 |
| Depravity | No. There is no content filter layered above the conversation, so there is nothing to switch off. | Not applicable. We do not generate images or video at all. Companion art is original cameo art and never a real face. | Three hard lines printed on every page: nothing involving minors, non-consent, or illegal activity. Nothing hidden underneath them. | Our own published policy |
| ChatGPT | Yes. The OpenAI Model Spec carries a rule titled "Don't respond with erotica or gore" at the System authority level. | Same rule covers text, images and audio. | Usage policy enforcement. No user prompt can reach the rule, because System sits above user and developer instructions. | model-spec.openai.com, five dated versions, read September 1, 2026 |
| Character.AI | Yes, and it is the strictest of this group on content while also running the strictest age assurance. | Filtered. | Open-ended chat was removed for under-18 accounts, with age assurance built in-house. | book.character.ai and blog.character.ai |
| SpicyChat | No. | On the app build, in the most literal way possible: the iOS app was pulled after Apple "determined that the app does not comply with their guidelines," and Android is a direct APK download rather than a Play listing. | Not documented in the pages we could read. | docs.spicychat.ai |
| Chub AI | No. Its own wording is "Chub is uncensored. We do not restrict in any meaningful capacity." | Payment processors, by its own account. On avatars it asks users to "avoid NSFW images whenever possible, it spooks the payment processors." | Not readable. chub.ai/terms returned a 403 bot block when we checked. | chub.ai, re-checked September 1, 2026 |
| Replika | Yes for most accounts. Erotic roleplay remains only for accounts created before February 1, 2023. | Filtered. | Terms clause 6.4: "If you are under 18, you are not authorized to use the Services." | replika.com terms, previously verified |
Kindroid quotations are verbatim from its public help center, read first-hand on September 2, 2026 at kindroid.ai. Those pages are rendered client-side, which is why they appear blank to many tools; we rendered them to read them. Pricing and policy in this market change often, so treat every row as accurate on the date shown rather than permanently. Two rows deserve emphasis because they cut against us: Kindroid explicitly permits consensual non-consent roleplay, which our hard lines do not allow, and Kindroid generates images and video, which we do not do at all.
Is Kindroid NSFW?
Yes, for text. Kindroid moderation guidelines contain a section headed "What is Explicitly Allowed," and the first item on it is "NSFW (Not Safe for Work) and Erotic Roleplay (ERP)." The stated reasoning is that "AI companions should be able to have the whole breadth of legal human adult experiences." Fictional violence, horror and darker creative themes are on the same list.
The FAQ puts it the same way from the other direction. Kindroid describes itself as "a neutrally aligned, unfiltered AI, because we believe this creates for more authentic interactions and that private use of AI should align with what the user wants and not be filtered by 3rd parties." It adds that the model "will not bring up any unethical content by itself, but its outputs are entirely dependent on what you input."
This is worth stating plainly because a large share of what is written about Kindroid hedges on it. Kindroid is not coy about adult content and does not bury the answer in a terms document. It publishes the permission, gives the reasoning, and then publishes the boundaries separately. Whatever you think of the product, that is a more honest posture than most of the market manages.
What is the NSFW toggle with Kindroid?
There is no NSFW toggle for chat, because there is no chat filter to turn off. People search for one because almost every other companion app has trained them to expect a settings switch. Kindroid architecture is different: the model is left alone and enforcement happens at the account level after the fact, so there is no per-message gate for a toggle to control.
A toggle does exist, but it is for media and it is only in one place. In the video selfie documentation Kindroid states that "you can choose to use the SFW or NSFW engine of videos on web only," and that "apps will only use the SFW engine." That is the switch people are half-remembering, and if you are looking for it inside the phone app you will never find it, because it is not there.
The same page explains the wider difference. "There is a keyword-based NSFW filter on words that may trigger NSFW images on the app, but not when Kindroid is accessed from the website." Kindroid gives the reason without being asked: "This is a precaution for certain app store policies dictated by Apple and Google." Then, flatly, "The filter does not exist on the web version of Kindroid."
Does Kindroid allow NSFW images and will Kindroid send explicit videos?
Kindroid generates both images and video, which it calls selfies and video selfies, and what you can get depends less on your subscription than on which door you came through. On the website there is no keyword filter on media prompts and the NSFW video engine is selectable. In the iOS and Android apps the keyword filter runs, and Kindroid says it "will be more robust for video selfies due to the existence of image input on apps."
The filter applies to prompts you write and to automatic selfie prompts alike. Your subscription does not change this. Kindroid is explicit that the surface, not the payment, is what decides: "Even if you subscribed on the app, you will have full access to the web version and vice versa," and the difference in filtering is the one thing that is not the same between them.
We should concede the obvious here rather than dance around it. Depravity does not generate images or video at all. If visual output is what you are shopping for, we are not the product, and no amount of comparison table can change that. What we offer instead is original cameo art for each companion, never a real face, and a conversation that does not stall.
What gets you banned on Kindroid?
Three things, named and worked through with examples. Kindroid calls them the Three Red Lines: imminent self-harm, imminent harm to others, and child sexual abuse material. The document is unusually careful about the boundary of each, giving paired examples of what does and does not count, and the distinction it keeps returning to is planning versus depicting.
On self-harm, venting that "I am so depressed, the world feels gray" is listed as not a violation, while stating a method and a timeline is. On harm to others, fictional violence and even a stated NSFW consensual non-consent scenario are listed as not violations, while using the AI to plan a real assault or to analyze a real person photos for stalking are. On CSAM the policy is absolute, minors are defined as under 18 "regardless of user jurisdiction," and it closes the obvious loophole in advance: "If a user labels a character as an adult while deliberately describing, portraying, or sexualizing them as a minor, we will treat the character as a minor."
The enforcement mechanics matter as much as the categories. Kindroid states that it "always issue a warning before locking an account," that warnings expire after two days, that scans cover current context only because "historical chats/media are not scanned," and that during automated detection "no human reads any content." A human only reads anything if you appeal, and only then with your explicit consent to decrypt the specific material that caused the lock.
How much does Kindroid cost?
Kindroid publishes real numbers, which is rarer in this market than it should be. Standard access is $13.99 a month, $37.99 for three months or $139.99 a year when bought directly on the web. Through the App Store or Google Play the same tiers are $15.99, $42.99 and $159.99. Kindroid explains the gap itself: prices are lower on the web "due to lower processing fees compared to app stores."
Above that sit two memory add-ons that most write-ups miss entirely. Ultra adds $24.99 a month on the web or $28.99 through an app, on top of the standard plan. MAX requires both and adds $59.99 web or $68.99 app, which puts a full MAX subscription at $98.97 a month on the web and $113.97 through an app store. Kindroid says these tiers exist to cover dedicated server costs and that it expects to "break even and potentially lose money even at these rates."
One clause is worth reading before you subscribe anywhere, including here. Kindroid states that refunds are considered case by case, that it "generally do not offer refunds and all sales are final," that it reserves "the right to permanently ban users who dispute charges," and that if an account is suspended or terminated for a terms violation you are "not entitled to any refund or credit for the unused portion of your subscription." Those two policies interact: the moderation system that locks accounts is the same system that can leave a paid year unrefunded.
Does Kindroid verify your age?
Yes, and its published approach is more specific than almost anything else we have audited in this market. Kindroid requires users to be over 18 and runs a two-track system. If you pay, your payment method is the verification, and Kindroid says that "once a user becomes paid, they are verified and will not need re-verification even if their subscription lapses." Free accounts instead pass through a behavioral age estimation system that "runs passively in the background."
If that system flags a free account, you choose between subscribing or verifying through Yoti, a third-party provider that runs either a selfie-based age estimate or a government ID check. Kindroid states that Yoti "deletes all images and biometric data after verification is complete" and that "Kindroid does not receive or store any images or identity documents."
Kindroid also published the outcome numbers, which is genuinely unusual. Fewer than 2% of new free accounts have triggered the estimation system. Of those flagged, roughly 55% never completed verification, which Kindroid reads as "consistent with these accounts belonging to individuals unable to confirm they meet the platform 18+ requirement." Among those who did verify, 80% used Yoti and 20% subscribed.
Kindroid alternative: how Depravity is different
The honest comparison starts with what Kindroid does better. It generates images and video, it runs group chats, it sells enormous memory tiers, and it permits some scenarios our hard lines do not, consensual non-consent roleplay among them. If any of those is your reason for shopping, Kindroid is the more capable product and you should buy it.
Where we differ is structural rather than a feature list. We never shipped a phone app, so no app store reviewer has a say in what your conversation is allowed to contain, and there is no second, quieter version of the product for people who found us through a browser. Kindroid documentation describes exactly that split; we simply do not have one to describe. Everything we offer behaves the same way because there is only one surface.
The rest is the ordinary product difference. We are conversation first: no image generation, no character marketplace, no public catalogue of other people creations. Companions have original cameo art and a personality you shape, memory that holds across sessions, and voice notes or voice calls on the higher tiers. Our three hard lines are printed on every page, and there is nothing underneath them.
Why the app stores keep deciding what adult AI can do
Read enough of these policies and the same shape appears everywhere. Kindroid says its media filter exists as "a precaution for certain app store policies dictated by Apple and Google." SpicyChat published a notice that its iOS app was removed because Apple "determined that the app does not comply with their guidelines," and it distributes on Android as a direct APK rather than through the Play Store. Chub AI asks users to avoid explicit avatars because "it spooks the payment processors."
None of those three companies changed their mind about adult content. Their distribution changed the product. The rule that reaches you as a filter, a missing toggle or a vanished app was usually written by a platform gatekeeper or a payment network, several steps upstream from the company whose logo is on the app.
That is the single most useful thing to carry away from all of this, and it makes a good test when you are comparing tools. Ask where a product is distributed before you ask what its content policy says. A browser-based service answers to its payment processor. An app store service answers to Apple, Google and its payment processor, and the strictest of those three sets the ceiling.
What you get
- No app store veto
- Limits published, all three
- No filter to toggle
- Original art, never a real face
It is ready to talk.
Start a chat and talk openly in about ten seconds. The demo is your free taste.
Good to know
Questions about your kindroid nsfw
Start now
Kindroid publishes exactly what it allows and exactly what gets an account locked. Almost nobody has read it, because the docs do not render without JavaScript. Try it above and see for yourself.
18+ · No filters, no lectures, no judgment. Private, always.