Panjabi Go Last updated: 4 June 2026
1. Why this policy exists
Panjabi Go is operated by Nihaal Group Ltd, a company registered in England and Wales. The app uses AI to help you practice Panjabi: a conversational AI tutor, AI-generated short stories, a translator, and other tools that produce text and audio on demand. AI is powerful but imperfect: it can sometimes produce content that is incorrect, inappropriate, or offensive. This policy describes what we don't allow, how we moderate, and how to report content that shouldn't be there. We have zero tolerance for objectionable content and abusive users.
2. AI-generated content disclosure
Any response produced by an AI feature in Panjabi Go is labeled as such in the app. Treat AI responses as a practice aid. They may contain mistakes, paraphrases, or unusual word choices. For authoritative language guidance consult a human teacher or verified dictionary.
3. Which features this covers
Content moderation applies to every AI-powered surface in the app, including:
- The AI voice chat and conversational AI tutor
- AI-generated short stories
- The translator, including text recognised from images you submit (OCR)
- Text-to-speech, where you supply the text to be spoken
- Open-answer assessment grading, where you supply an answer
4. What's not allowed
You must not use the AI features to request, generate, or share content that falls into any of the following categories:
- Sexual content or solicitation
- Content that sexualizes or endangers minors
- Hate speech, harassment, or content targeting protected groups
- Instructions or promotion of self-harm, suicide, or eating disorders
- Credible threats, glorification, or detailed planning of violence
- Instructions for making weapons, including firearms and explosives
- Illicit drug sale or manufacturing instructions
- Content that impersonates real individuals in a harmful or deceptive way
- Attempts to extract personal or private information about others
We may also block or remove content that isn't explicitly listed here but violates the spirit of this policy. We reserve the right to update the list as our AI features evolve.
5. How we moderate
Every message sent to an AI feature is checked against an automated content moderation classifier (OpenAI's Moderation API) in two places:
- Your input is checked before the AI sees it. If it's flagged, we refuse the request and show a brief notice.
- The AI's output is checked before you see it. If it's flagged, we drop the output and substitute a generic response.
For short stories, each story is automatically moderated before it becomes visible to readers. Stories that are flagged are not published.
6. Consequences for misuse
Accounts that repeatedly trigger moderation flags are automatically rate-limited and may be escalated for human review:
- 3 flagged inputs within 24 hours → AI features are temporarily paused for that account for one hour.
- 10 flagged inputs within 7 days → the account is flagged for review by our safety team.
In serious cases, access to the AI features may be suspended or the account terminated. Our safety team reviews reports and high-severity events on an ongoing basis.
7. Reporting content
Every AI-generated item in the app has a flag icon. Tap it, pick a reason, and add optional detail. After you submit:
- The reported item is hidden immediately on your device.
- The report is sent to our safety team in real time.
- We act on egregious or illegal content within 24 hours, and aim to triage all other reports within 72 hours.
You can also block content and email us directly at support@panjabigo.com if you prefer.
8. What we store
For each moderation check we keep an audit record containing a one-way SHA-256 hash of the checked text (not the text itself), along with the moderation result and timestamp. This lets us investigate reports and enforce the rules above without retaining the raw content. AI conversations and pronunciation audio you create are stored as described in our Privacy Policy.
9. Appeals
If your account has been rate-limited or restricted and you believe this was in error, reply to the notice in-app or email support@panjabigo.com. We'll review the events on record and respond.
10. Changes to this policy
We'll update the "Last updated" date at the top of this page whenever the policy changes materially. Continued use of the AI features after an update means you accept the updated policy.
