These terms govern the use of the artificial intelligence (AI) powered features within Pray Assist and form an integral part of the Terms of Use. By using the AI features, you accept these terms.
1. Purpose and Scope
Pray Assist offers AI-powered features such as the AI guide that answers your questions about worship, recitation feedback, voice-message transcription, text-to-speech (TTS), reading ingredient labels in the halal checker, inappropriate-content detection in lessons, lesson suggestions, and profanity/communication filtering. These terms cover only these features.
2. Nature of AI Output and Human Oversight
The AI features are an assistive, probabilistic decision-support system. Answers, suggestions, halal assessments, moderation, and transcription results may be inaccurate, incomplete, or misleading; they are not binding, professional, legal, medical or religious rulings (fatwa), nor established fact. Final judgment and responsibility rest with you; do not base important decisions solely on AI output.
3. Acceptable Use
You may not use the AI features to produce unlawful, hateful, harassing or sexual content, to imitate a religious authority or to present an AI answer as a ruling issued by a named scholar, to extract another person's personal data, or to attack the system through automated requests, prompt injection or attempts to bypass its safeguards. Accounts breaching this section may have their access to the AI features restricted or closed.
4. Data Processing, Model Training and Third Parties
The Company may process data under anonymization and isolation principles to improve the accuracy of the features. Some AI features run through the enterprise APIs of third-party providers: Anthropic (Claude), OpenAI, Google, and AWS. Enterprise / "Zero Data Retention" and no-training principles are observed with these providers. The Company is not liable for unilateral changes or breaches by third parties.
5. Intellectual Property
You retain rights to your inputs and the AI outputs you receive. The underlying infrastructure — algorithms, model weights, prompts, architecture, and software — belongs to the Company. Reverse engineering and using outputs to train a competing model are prohibited.
6. Availability and Quotas
The AI features may be subject to usage quotas that differ by subscription tier, and may be interrupted, limited or withdrawn because of provider outages, cost, abuse or legal requirements. Continuous availability is not guaranteed and a quota does not constitute a separately purchased entitlement.
7. Limitation of Liability
The AI features are provided "as is"; the accuracy, uninterrupted operation, or fitness for a particular purpose of the output is not guaranteed. The Company is not liable for indirect or incidental damages, data loss, or consequences arising from reliance on the output. These terms may be updated from time to time.
8. Contact
Contact: [e-mail]
Last Update: 2026-08-10