Jada AI — Transparency & Policy

1. What Jada Is

Jada is an AI assistant powered by large language models (LLMs) integrated into the Jada platform. Jada is designed to:

• Help Event Planners receive personalised vendor recommendations, explore event ideas, build shortlists, and navigate the planning process.
• Help Vendors write business bios, generate custom FAQs, develop pricing strategy, draft professional proposals, analyse technical riders, and improve their profile and communication.
• Provide contextual suggestions and content generation based on information you share within the platform.

Jada is a tool to assist and inform — it does not make decisions on your behalf, enter into contracts, or guarantee any outcome.

2. What Jada Is Not

• Not a professional advisor. Jada’s outputs do not constitute legal, financial, insurance, tax, or event management advice. Always seek qualified professional advice for those matters.
• Not infallible. Like all AI systems, Jada can produce inaccurate, incomplete, outdated, or inappropriate outputs. This is commonly referred to as “hallucination. ” You should review and verify any content Jada generates before acting on it or using it publicly.
• Not a verification service. Jada does not verify vendor credentials, licences, insurance, reviews, or the accuracy of vendor-provided information.
• Not a decision-maker. Jada does not decide which vendor you hire, what contract terms you agree to, or what price you charge. All decisions remain entirely with you.
• Not a real-time data source. Jada does not have access to live vendor availability, live pricing, or real-time external information unless explicitly integrated.

3. Data Jada Uses to Generate Responses

For Event Planners

• Event details you have entered: type, date, location, guest count, budget, style preferences, and notes.
• Your shortlist activity and search behaviour within the current session.
• Vendor data from the Jada platform: category, location, pricing range, and profile completeness — used to generate relevant recommendations.

For Vendors

• Your business profile: name, service category, description, location, pricing, and completion status.
• Content you share within the Jada chat — such as text you want rewritten, a proposed bio, or pricing questions.
• Files you upload (e.g., PDF technical riders) for analysis features, where applicable.
• Lead and event data visible in your vendor dashboard, where context is needed for lead-specific assistance.

Jada takes care to minimise the personal data passed to AI processing. We do not pass sensitive personal data (such as payment details or government ID) to Jada.

4. How Conversation Data Is Handled

• Processing: messages you send to Jada are processed in real time to generate responses. This processing may involve third-party AI infrastructure providers (such as Anthropic or OpenAI) under data processing agreements that prohibit them from using Jada user data to train their own public models.
• Retention: conversation logs may be retained for up to 12 months to allow you to review your chat history and to enable us to improve the service. You may request deletion at any time (see Section 7).
• Isolation: your conversation data is not shared with other users of the platform. Vendors cannot see Planner conversations and vice versa.
• Service improvement: Jada may use aggregated, anonymised conversation patterns (never individually identifiable messages) to improve Jada’s performance and accuracy.
• No sale of data: conversation data is never sold or shared with third parties for their own commercial purposes.

5. No Binding Automated Decisions

Jada does not use Jada to make legally binding or significantly consequential decisions about you solely through automated means. Specifically:

• Vendor recommendations are suggestions only — Planners choose independently.
• No access to the platform, account status, credit allocation, or lead distribution is determined exclusively by automated processing without human involvement.
• You retain full control over all hiring, booking, and transactional decisions.

Your rights under PIPEDA and GDPR to human review of automated decisions are respected. If you believe an automated process has affected you in a significant way, contact us at contact@eepic.ca.

6. Limitations & Known Risks

• Hallucination: AI-generated content may contain factual errors or invented information presented confidently. Always verify critical details independently.
• Bias: language models may reflect biases present in their training data. Jada monitors for harmful outputs and works to mitigate bias, but cannot guarantee that all outputs are free from it.
• No credential verification: Jada does not confirm that vendors hold required business licences, professional certifications, or adequate insurance.
• Pricing guidance is general: pricing suggestions from Jada are for general guidance only and do not account for your specific market conditions, costs, or legal obligations.
• PDF analysis is best-effort: technical rider and document analysis is a best-effort feature. Critical details in uploaded documents must be verified directly with the client.
• Marketing content requires review: any AI-generated text for public use (bios, FAQs, proposals) should be reviewed and edited by you before publishing. You are responsible for the accuracy and appropriateness of any content published on your profile.

7. Your Rights Regarding AI Data

You have the following rights in relation to data processed by or through Jada:

• Deletion of conversation history: email contact@eepic.ca with subject line “AI Data Deletion” and we will delete your Jada conversation logs within 30 days.
• Access to AI data: you may request a summary of what data was used in your Jada interactions by emailing contact@eepic.ca.
• Opt out of service improvement use: if you do not want your anonymised conversation patterns used for service improvement, email us with subject “AI Improvement Opt-Out”. Note that opting out does not affect core Jada functionality.
• Report harmful outputs: if Jada produces content that is inaccurate, harmful, biased, or offensive, please report it (see Section 9).

All other privacy rights under PIPEDA apply — see our Privacy Policy for full details.

8. Regulatory Context

• Canada — AIDA (Bill C-27): Canada’s Artificial Intelligence and Data Act is currently in the legislative process. Jada is monitoring its requirements and will update our practices as the law comes into force.
• EU AI Act: the EU AI Act (in effect 2024) classifies AI systems by risk level. Jada’s use of AI for event vendor recommendations is considered a low-risk application. Jada does not use AI for any high-risk purposes as defined by the EU AI Act (e.g., credit scoring, employment decisions, biometric identification, or law enforcement).
• PIPEDA & Quebec Law 25: Jada’s data processing is governed by our Privacy Policy and complies with PIPEDA’s principles of purpose limitation, data minimisation, and individual access rights.
• GDPR: for users in the EU/UK, we rely on legitimate interests and, where applicable, consent for AI data processing. No solely automated decision-making with legal effect is performed (Article 22 GDPR).

9. Human Oversight

Jada maintains human review processes for reports of harmful, inaccurate, or misleading AI outputs:

• Report a problem: email contact@eepic.ca with subject “Jada Feedback” and describe the issue, including the output you received.
• We review all reports and may modify, retrain, or restrict Jada features in response to identified issues.
• Critical safety-relevant reports (e.g., outputs that could cause harm) are escalated and addressed as a priority.

10. Updates to This Policy

We may update this policy as Jada’s capabilities evolve or as AI regulations develop. For material changes, we will provide at least 14 days’ notice via in-app notification or email. The “Last updated” date at the top reflects the most recent revision.

11. Contact

Questions about Jada or this policy? contact@eepic.ca