Trust and safety

Automation & AI transparency

What BubsBookings’s Safety Bot does, what data it uses, and where automated judgment stops.

Effective September 3, 2026

What the Safety Bot checks

The BubsBookings Safety Bot checks newly submitted chat messages, service titles and descriptions, booking notes, cancellation reasons, and reviews. It looks for profanity, harassment, hate, explicit sexual content, exploitation, credible threats, and language associated with serious illegal activity.

What happens when content is detected

Content that matches a safety rule is not posted. The user receives a clear request to revise it. BubsBookings records a limited event containing the account, feature, safety category, severity, time, and a one-way fingerprint so repeated abuse can be investigated without storing a separate copy of the blocked text.

Technology and attribution

The Safety Bot is a BubsBookings-operated automated rules system. It does not generate messages, make booking decisions, determine prices, or rank providers, and it does not send private chats or booking text to a generative-AI provider for moderation.

BubsBookings AI is a separate, optional help assistant for eligible paid provider accounts. When a user submits a question, BubsBookings sends that question and the recent assistant conversation to OpenAI’s API to generate an answer. BubsBookings instructs the assistant not to request sensitive information, and requests are sent with API response storage disabled. BubsBookings AI can make mistakes and cannot change bookings or accounts.

BubsBookings AI is powered by OpenAI. BubsBookings’s transparency and risk-documentation approach is informed by the voluntary NIST AI Risk Management Framework.

Limits

Automated moderation can miss harmful language or block harmless language, especially slang and context-dependent phrases. A passed check is not an endorsement, safety guarantee, background check, or substitute for human judgment.

Safety and emergencies

Do not rely on BubsBookings for emergencies. If someone may be in immediate danger, contact local emergency services. Use account security controls if you suspect unauthorized access, and stop communicating with users who pressure you to leave the platform or share sensitive information.

Fairness and accountability

BubsBookings applies the same text rules to customers and providers. Safety checks do not use protected characteristics to approve bookings or rank people. BubsBookings will document material changes to the system and provide a support path for users who believe content was blocked incorrectly.

These terms are a practical launch policy and should be reviewed by a qualified attorney before BubsBookings accepts payments or expands into additional jurisdictions.