AI-generated replies
Useful when a question needs a contextual response. Cloud generation involves external AI processing, and a generated answer can still be incomplete or incorrect.
Bujji AI brings AI-generated replies and configured response rules to WhatsApp Web. Choose the approach that suits the conversation, understand what context is used and keep control over when automatic replies are enabled.
An AI-generated reply uses a model to produce an answer from the prompt and the context supplied to it. A keyword reply follows a configured rule when a message matches that rule. Bujji AI has separate feature controls for these approaches; enabling one should not be treated as proof that every reply uses the other.
Useful when a question needs a contextual response. Cloud generation involves external AI processing, and a generated answer can still be incomplete or incorrect.
Useful for repeatable responses to matching words or conditions. Review the rule and its intended scope before letting it handle incoming messages.
Choose when automatic replies are enabled. Disable the feature to stop automatic responding, and test the intended behavior before using it for customer conversations.
Enterprise Brain is Bujji AI’s knowledge-based assistance workflow. It can retrieve relevant information from supported business files that you directly upload or from information you enter. Selected retrieved fields may be included in a cloud prompt to help generate a reply.
Adding knowledge does not train a new foundation model, guarantee an accurate answer or remove the need for human judgment. Keep your source information current, check representative questions and test what happens when the information needed to answer is missing.
Learn more about Enterprise Brain and the knowledge-base data boundary.
Decide whether the task needs a fixed response rule or AI generation. Confirm feature access and the controls exposed by the released version.
Use only information you are allowed to provide. Understand which conversation excerpts and selected business context may be sent for cloud generation.
Use controlled test conversations first. Check a matching request, an unrelated question and the action that disables automatic replies.
Message matching and configured automation rules run on your device. When a reply requires cloud AI, prompt text, relevant message excerpts and selected knowledge context are processed through Bujji AI’s relay infrastructure by the service handling that request.
Provider processing and retention depend on the applicable service and configuration. The product privacy policy does not promise universal zero retention. Do not describe cloud-generated replies as fully offline, and avoid supplying sensitive information without understanding the processing path.
Bujji AI states that Google user data is not transferred to AI providers. Directly uploaded business files and WhatsApp conversation context are different sources from data obtained through Google APIs. See the privacy policy for the full explanation.
The browser workflow depends on WhatsApp Web being available. The homepage describes an optional Desktop Engine for supported planned messaging when the browser is closed; that does not establish that every AI reply mode operates independently of Chrome.
Confirm the requirements for the specific AI reply mode in the released product. Network availability, WhatsApp connection, feature access and the current device setup can affect operation. This guide does not promise uninterrupted replies, guaranteed delivery or protection from account restrictions.
For a knowledge-based response, relevant catalog information can supply context to the AI generation step. If the catalog does not contain the answer, the operator should not assume the generated response is correct. Review uncertain or consequential answers and keep a way to stop automation.
No. Keyword replies follow configured matching rules. AI replies use model generation and can use selected context. Choose the workflow that matches the conversation and review the configured behavior before enabling automatic replies.
Cloud-generated replies are not fully offline. Local matching and retrieval are separate from external model processing. Read the privacy policy to understand what text and context may leave the device.
The privacy policy states that you can stop automatic replies by disabling the feature. Test that control with your intended setup before enabling it for live conversations.
Do not assume that every AI reply mode works with Chrome closed. The optional Desktop Engine has its own supported workflows and prerequisites. Check current mode-specific release instructions.
AI-generated replies and keyword replies have different feature access rules. Confirm the published plan details at launch; check the homepage for current availability, prices and usage limits.