Emergency and Safety Disclaimer

Document Type

Safety Notice

 Version

1.1

 Effective Date

August 27, 2026

 

Important Safety Notice

Frontline Connect is an AI-assisted wellness, engagement, and early-support platform. It is not a crisis hotline, emergency response service, substitute for professional diagnosis or treatment, or a continuously monitored clinical service.

What Frontline Connect Is — and Is Not

This section states plainly what the platform can and cannot do. It governs the description of the platform’s role in every other Frontline Connect document, including the Participant Consent and Agreement, Terms of Use, Privacy Policy, and Research Informed Consent.

Frontline Connect Is:

  • An AI-assisted engagement and wellness tool that offers structured check-ins, reflection prompts, and conversational support.
  • A safety-screening layer that reviews check-in content for signs of elevated risk and can trigger a human alert.
  • A resource-navigation service that can connect participants to peer, community, clinical, or crisis resources.
  • A support surface with human involvement in the sense that authorized personnel may review flagged content and reach out — but that outreach is not clinical treatment (see below).
  • A platform with administrator access and monitoring capability. Application administrators have full access to chat logs and can monitor conversations, including in real time, for safety, security, and operational purposes (see “Administrator Access and Monitoring” below).

Frontline Connect Is Not:

  • An emergency service or a replacement for 911, 988, or a local emergency department.
  • A crisis hotline or a real-time crisis intervention service.
  • A provider of medical or psychiatric diagnosis, treatment, or psychotherapy. Nothing in a check-in, an AI response, or a follow-up contact from support personnel constitutes a clinical evaluation, diagnosis, or treatment plan.
  • A guaranteed clinical response, even though administrators have real-time access to chat logs. Administrative visibility is a monitoring and safety capability, not a guarantee that any particular check-in is reviewed by a clinician or that any particular response time will be met.
  • A guarantee of a human response. Automated classification can be wrong in either direction, and staff availability is not continuous (see below).

Administrator Access and Monitoring

In addition to the automated review described below, designated Frontline Connect application administrators have full access to participant chat logs and can monitor conversations, including in real time. This access exists to support platform safety, security, quality assurance, and operational oversight.

  • Scope of access. Administrators with this access can view the complete content of a participant’s check-ins and conversations with the AI system, not only content flagged as elevated risk.
  • Real-time capability, not a guarantee of real-time review. That administrators can monitor conversations as they happen does not mean every conversation is actively watched at all times. Ongoing review is prioritized based on risk classification and staffing, as described above.
  • Not a substitute for calling 911 or 988. Participants should never rely on the possibility of administrator monitoring in place of contacting emergency services directly.

If You Need Immediate Help

  • Call 911 or go to the nearest emergency department if you or another person is in immediate danger.
  • Call or text 988 to reach the Suicide & Crisis Lifeline. Veterans may press 1 to reach the Veterans Crisis Line.
  • Do not wait for a response from Frontline Connect, a peer supporter, or a clinician when immediate intervention is needed.

AI Review and Human Follow-Up

Frontline Connect uses automated systems to review check-in content for potential safety concerns. If a message is classified as elevated concern, authorized support personnel may receive an alert containing identifying information and relevant portions of the check-in. A support team member may review the check-in and contact the participant to offer support, confirm safety, or assist with connection to appropriate resources.

Automated classification and human outreach are safety supports, not guarantees. Technology may misclassify content, alerts may be delayed, and a responder may not be immediately available.

The Limits of Human Follow-Up

Participants should understand the practical limits of this safety layer so expectations are accurate:

  • Response is not immediate. Alerts are reviewed by staff during their working hours, not by an always-on clinical team.
  • Response is not guaranteed. Staff availability, technical issues, or classification errors may mean an elevated-risk check-in receives no follow-up, or delayed follow-up.
  • Outreach is not treatment. A support team member’s contact is a wellness check and resource connection — it is not psychotherapy, a clinical evaluation, or a substitute for emergency services.
  • Support personnel are not necessarily the participant’s treating clinician. They may be peer supporters, clinicians, or other designated responders, depending on staffing and program design.

Participant Acknowledgment

By using Frontline Connect, the participant acknowledges that:

  • the platform is not an emergency service and is not a substitute for medical or psychiatric diagnosis, treatment, or crisis intervention;
  • they understand when to use 911 or 988;
  • AI systems review messages, and this review is a safety support rather than a guarantee;
  • authorized support personnel may review and respond to elevated-risk check-ins, and that this outreach is not guaranteed, immediate, or a form of clinical treatment;
  • Frontline Connect administrators have full access to chat logs and can monitor conversations, including in real time, for safety, security, and operational purposes.

Frontline Connect | Massachusetts Veterans Suicide Prevention Institute | This Emergency and Safety Disclaimer is the governing source for the platform’s scope and limitations and is incorporated by reference into the Participant Consent and Agreement, Terms of Use, Privacy Policy, and Research Informed Consent.