Service Level Agreement

• "Uptime" means the percentage of time the applicable service is available for use during the measurement period, excluding exclusions defined below. • "Scheduled maintenance" means maintenance communicated at least 48 hours in advance (or 72 hours for major releases where possible). • "Incident" means an unplanned interruption or reduction in quality of the service. • "Response time" means the time from when we acknowledge your support request to when we begin active work or provide an initial response.

We target 99.5% uptime for the core INITE AI platform (excluding exclusions) measured monthly. Uptime is calculated as: (total minutes in the month minus excluded minutes minus downtime minutes) divided by (total minutes in the month minus excluded minutes), expressed as a percentage. If we fail to meet this commitment in a given month, you may be eligible for service credits as described in the "Service credits" section, subject to your timely claim and continued paid subscription.

The following are excluded from uptime and response-time calculations: • Scheduled maintenance (with advance notice). • Issues caused by your or your users' equipment, network, or actions (e.g. misuse, misconfiguration). • Force majeure events (natural disaster, war, terrorism, pandemic, government action, major third-party provider failure beyond our control). • Your breach of the Terms of Service or Acceptable Use Policy. • Attacks or incidents that we mitigate in accordance with industry practices (e.g. DDoS mitigation).

We aim to respond to support requests according to severity: • Critical (production down or major feature unavailable): initial response within 4 business hours. • High (significant degradation, workaround available): initial response within 8 business hours. • Medium (partial issue or non-critical): initial response within 24 business hours. • Low (questions, minor issues): initial response within 48 business hours. Business hours are defined in your region or as stated in your support plan. Response time is not a guarantee of resolution time; we will work toward resolution based on severity and resource availability.

Support is provided via email or the support channel specified in your plan. For critical issues, we may provide a designated contact or escalation path as per your agreement. You must provide sufficient detail (description, steps to reproduce, impact) to allow us to investigate. We may request additional information or access (under appropriate confidentiality) to resolve issues.

We will use reasonable efforts to perform maintenance during low-usage windows and to give at least 48 hours' notice (or 72 hours for major changes when possible). Emergency maintenance may be performed with shorter or no notice when necessary to protect security or stability. We will post status updates where feasible.

If we do not meet the uptime commitment in a given month and you submit a valid claim within 30 days of the end of that month, we may issue a service credit as a percentage of your monthly fee for the affected service, in accordance with our then-current credit table (e.g. 99.0%–99.5%: 10% credit; 95.0%–99.0%: 25% credit; below 95%: 50% credit). Credits are applied to future invoices and are your sole remedy for SLA failures unless otherwise required by law. You must be in good standing (paid, no material breach) to receive credits.

To claim a service credit, contact us with the subject "SLA Credit Claim" and include: (1) your account or order reference, (2) the month and service affected, (3) a brief description of the incident and impact, and (4) any ticket or incident reference if applicable. We will review and respond within 15 business days. Disputes will be resolved in accordance with our Terms of Service.