← sry.fyi
Privacy Policy
Last updated: 2026-07-16
The short version. Sry is privacy-light:
- No account, no ads, no tracking for advertising.
- We do not sell your data, ever.
- We do not keep the apologies you generate in our long-term database (a short-lived playback cache holds them for about 30 minutes, then they expire).
- To run the app, the text you enter and the apology we generate are processed by third-party service providers (currently OpenAI for AI text and audio, plus Cloudflare and Apple), described below.
1. Who we are
Sry (also marketed as "Sry: Apology Slider") is a comedy mobile app and the accompanying website at sry.fyi. Reach us at hello@sry.fyi.
2. What we collect and process
To run the app and prevent abuse, we collect or process only the following:
- A daily rate-limit counter per device. A small integer that counts how many apologies you've generated today on this device. Used solely to enforce per-device daily limits (3 generations/day for the free tier, a higher fair-use cap for paid). The counter resets at the end of each UTC day and is automatically deleted after 25 hours. We use Apple's Identifier for Vendor (IDFV) as the key, which is scoped to our app and resets if you uninstall. We do not use this identifier for advertising or for tracking you across other companies' apps or websites. It is not Apple's advertising identifier (IDFA), and the app does not request tracking permission.
- Anonymous, aggregate usage counts. Tallies like "5,000 apologies generated today" or "1,200 of those used the Boomer Text persona." These counts have no device-level breakdown and contain no content of your apologies. We use them to understand which personas and intensity levels are popular, so we can decide what to build next. Auto-deleted after 60 days.
- The optional context you type, processed at request time. Whatever you enter in the "what happened?" field (e.g., "missed dinner") is sent to our backend and forwarded to our third-party AI text-generation provider (currently OpenAI) so the model can write a relevant apology. We do not keep this input text in our long-term database after the request completes. Provider-side processing is described in section 4.
- A short-lived playback cache. When you tap Play, the generated apology text, its spoken-audio file, and short-lived word-timing metadata (used to sync the on-screen captions) may be cached temporarily on our backend for about 30 minutes so replaying is fast and we don't re-bill the audio model. Because the word-timing metadata maps each word to a moment in the audio, this cache does contain the literal words of the generated apology text for that short window. The cache entry then automatically expires and is deleted. The cache is keyed by a one-way hash and is not linked to your device.
- Technical logs. Like most online services, our hosting and security providers process limited technical information such as IP address, device or browser type, timestamps, request metadata, and error logs. We use this only to operate the service, prevent abuse, secure the app, and diagnose problems. We do not use it for advertising.
3. What we do NOT collect
- Your name, email, phone number, or any personal account details
- The text of your generated apologies in our long-term database (a short-lived playback cache holds them for about 30 minutes, then they auto-expire, as described in section 2)
- Photos, contacts, location, microphone, calendar, or any other device data
- Apple's advertising identifier (IDFA); we do not run ads
- Crash reports linked to your identity (Apple's anonymized crash reports may be visible to us via App Store Connect)
4. Service providers
We use these services to operate the app:
- A third-party AI text and audio generation provider (currently OpenAI) to produce the apology text and spoken audio. The context you type and the persona prompt are sent to this provider so it can generate the apology text, and the generated apology text is then sent to produce the spoken audio. When you create a share video, the generated audio is also sent to the same provider's speech-to-text model (OpenAI Whisper) to calculate the word-by-word caption timing. OpenAI states that API data is not used to train its models by default. OpenAI may retain abuse-monitoring logs for a limited period under its standard data controls. See OpenAI's privacy policy.
- A hosting and content-delivery provider (currently Cloudflare) to serve our backend and website. Cloudflare automatically logs request metadata (such as IP, timestamp, and country) for security and abuse prevention.
- Apple, for app distribution, in-app purchases, and anonymized crash reports. Apple's privacy practices apply.
We do not use any third-party advertising or analytics SDKs.
5. International data processing
We only use reputable service providers (currently OpenAI, Cloudflare, and Apple) and rely on them to provide appropriate, industry-standard protection for the limited data they process on our behalf. Our service providers may process data in countries other than where you live. By using the app, you understand that limited data needed to operate the Service may be processed internationally, subject to those providers' security and privacy practices.
6. Data retention
- Daily rate-limit counters: 25 hours, then auto-deleted
- Anonymous usage counts: 60 days, then auto-deleted
- Playback cache (generated text, audio, and word-timing metadata): about 30 minutes, then auto-deleted
- Cloudflare request logs: per Cloudflare's standard retention
- OpenAI request data: per OpenAI's API data controls and retention
7. Your rights
Because we do not collect personal account data, there is no "account" to delete. To remove all data tied to your device, uninstall the app: the device counter resets, and the aggregate counts expire on their own schedule.
If you have any privacy questions or requests, email us at hello@sry.fyi. We aim to respond within 7 days.
8. Children and minors
Sry is rated 17+ for crude humor and is not intended for use by anyone under 17. We do not knowingly collect personal data from anyone under 17. If you believe a minor has used the app and any personal data needs to be addressed, contact us at hello@sry.fyi and we will take reasonable steps.
9. Security
All requests between the app and our servers are made over HTTPS. We do not store payment details, passwords, or any other credentials. If a security incident affects data we hold, we will take reasonable steps to notify affected users or post a notice as required by applicable law.
10. Changes to this policy
We may update this policy as the app evolves. Material changes will be reflected in the "Last updated" date at the top. Continued use of the app after a change constitutes acceptance.
11. Contact
Questions, concerns, or privacy requests: hello@sry.fyi.