AI Content Disclaimer

The following Fulcrum features generate content using AI models:

Talk/Xpress: voice call scripts, real-time conversation responses, voicemail transcripts.

FlowDesk: email sequences, lead outreach messages, quote generation, pipeline notes.

Responder: chat responses across WhatsApp, Facebook, SMS, web chat, and Google Business Chat.

Presence: social media posts for Facebook, Instagram, LinkedIn, and Twitter/X.

Pilot: knowledge base synthesis, document summaries, answer generation.

ScanPro: property listing descriptions, outreach templates.

All AI-generated content must be reviewed and approved by a human before it is published, sent, or acted upon.

The Service includes human-in-the-loop approval workflows for outbound actions. Do not bypass or disable these review mechanisms.

AI output is provided without any warranty. Specifically:

Accuracy: AI models may produce factual errors, incorrect data, or plausible-sounding but false information (hallucinations).

Completeness: AI output may be incomplete or omit relevant information.

Fitness for purpose: we do not warrant that AI output is suitable for any particular business purpose.

Bias: AI models may reflect biases present in training data, potentially producing discriminatory or skewed content.

Consistency: the same input may produce different output across requests.

AI output must not be used as a substitute for professional advice without review by a qualified human professional:

Client agrees to indemnify and hold harmless Fulcrum from any claims, damages, or liabilities arising from AI-generated content that the Client publishes, sends, or acts upon without adequate human review.

AI output may incorporate or be derived from third-party data or training datasets. Fulcrum does not guarantee that AI output is free from third-party intellectual property claims. Client is responsible for verifying that published content does not infringe third-party rights.

Information you provide to AI features (knowledge base documents, customer queries, business data) is processed by our AI sub-processors (Mistral AI). Do not input special category data (health, biometric, racial/ethnic, religious, political, trade union, sex life, criminal records) into AI features. See the Privacy Policy for details.

AI models are updated periodically. Output quality, style, and accuracy may change without notice. We do not guarantee backward compatibility of AI output across model versions.

Questions about this disclaimer: [email protected].

Incorporated into our Terms of Service by reference.