Terms of service

Last updated October 27, 2026

By using Tiny Desk AI you agree to these terms. They're short because the service is simple: you add photos, we make a parody video.

What the service is

Tiny Desk AI is an independent fan tool for making AI parody videos of the viral "Outstanding" Tiny Desk clip, for personal, non-commercial sharing. It is not affiliated with or endorsed by NPR, Tiny Desk Concerts, Charlie Wilson or The Gap Band. Tiny Desk is a trademark of NPR. The concert footage and the song stay with their rights holders, and making a parody doesn't move any of those rights to you, so don't sell your video or use it in ads. When a rights holder asks us to take something down, we do; see our DMCA page.

Your photos

• Public figures such as coaches, players and celebrities are fine for parody, as long as the video does not put words in their mouth, present them as saying something they did not, or is used to deceive, harass or embarrass them.
• No one under 18, and no photos of private people who wouldn't want to be in it.
• You keep any rights you have in your photos. You let us use them only to make your video.

How you can use the videos

• Personal, non-commercial parody: post it, send it to the group chat, play it at the party.
• Don't present it as real, use it to harass or embarrass someone, or put it in ads or endorsements.
• Platforms may mute or remove videos that use the song. That's between you and the platform.

Credits and payment

Videos are paid with credits from one-time packs. Credits don't expire and can't be exchanged for cash except as described in the refund policy. Prices are shown before you pay.

AI results

The video model aims for a strong likeness, not an exact copy, and results vary with the photo. The model may refuse some photos; a refused video returns its credits.

Acceptable use policy: what you cannot generate

This service puts a face into a parody music video, so we have to be strict about what can be made with it. The seven categories below are never allowed, including jokes, parodies and “it’s only for my friends”:

1. Sexual or NSFW content: pornography, explicit sexual acts, fetish content, sexual exploitation or explicit nudity.
2. Violence and gore: graphic injury, blood, glorification of violence, threats, or instructions that help someone commit violent wrongdoing.
3. Hate speech: content that attacks, dehumanizes, threatens or encourages discrimination against a person or a protected group.
4. Child sexual abuse material (CSAM) and any sexualised content involving a minor. There is no exception for fictional, stylised or “artistic” requests involving minors.
5. Non-consensual intimate imagery (NCII): generating, uploading or sharing intimate or sexual content depicting an identifiable person without their documented consent — including real footage, AI-generated or face-swapped versions of it, “revenge porn”, and leaked or voyeuristic private media. Judgement: content is NCII whenever it depicts an identifiable person in an intimate or sexual context and that person's consent cannot be established; AI synthesis, stylisation and “it was just a joke” are no exception. Confirmed NCII is removed immediately, the account that generated or shared it is suspended (permanently for repeat or malicious cases), and evidence is preserved and handed to law enforcement where required.
6. Deepfakes and impersonation: putting a real person’s face into a video in a way that misrepresents what they said or did, or that is used to deceive, harass, extort or embarrass them.
7. Copyright and trademark infringement: using protected work, music, logos or brand identity without the rights to do so, or removing rights-management information.

Other things you may not do

• Generate content that is illegal, defamatory, harassing or fraudulent.
• Produce malware, phishing content or cyberattack tools.
• Circumvent or interfere with security, moderation, billing or credit systems, or resell access to the service without our written approval.
• Violate any applicable law, regulation or card network rule.

AI-specific rules

• Use our inputs, outputs or model responses to train, fine-tune, benchmark or distil any model that competes with this service.
• Systematically scrape or harvest outputs at scale for your own use.
• Present AI output as the work of a human professional (a doctor, lawyer, financial adviser) in a context where that could cause harm.

Moderation, reporting and enforcement

We screen requests before generation where a safety screening service is configured, and the model may refuse on its own. A refused or failed generation does not use up your credits. We may refuse a request, remove content, restrict or close an account, revoke unused credits or decline future transactions. If you see prohibited content, or want to report a video that breaks these rules, email us and we will review it. General content reports are acknowledged within 1 business day and actioned — including removal where warranted — within 3 business days of receipt, with the outcome sent to the reporter; copyright notices follow the timeline in the DMCA section below.

When we can say no

We may refuse a video, remove content, or close an account that breaks these terms, and we remove content when a valid takedown notice arrives (see DMCA).

No warranty

The service is provided as is. To the extent the law allows, our total liability to you is limited to what you paid us in the last 12 months.

Changes

If we change these terms, we update the date at the top. Using the service after that means you accept them.

Questions? Email larrygofast89@gmail.com.