Effective date: May 2026
Last updated: May 2026
ATLAS helps with operational decision support, but human review remains essential.
ATLAS helps with operational decision support, summaries, recommendations, drafts, explanations, and workflow assistance.
Outputs may be wrong, incomplete, outdated, unsuitable, or based on incomplete records. AI may misunderstand data and may depend on provider behavior and available source records.
AI outputs are not guaranteed to be explainable. Citations, source links, confidence cues, and visual explanations may be incomplete, unavailable, or based on limited system context.
Humans must review decisions. Final responsibility stays with the customer and authorized users.
AI should not be used as the sole source for legal, medical, safety, financial, compliance, customer approval, or operational release decisions.
ATLAS is not an attorney, accountant, engineer, safety inspector, compliance officer, medical provider, or licensed professional.
ATLAS may summarize, recommend, draft, rank, prepare, or explain. Human approval is required for final financial, customer, safety, legal, or operational release decisions unless separately configured bounded automation exists.
Prompts, responses, tool calls, source references, approvals, and ATLAS action logs may be retained for accountability, audit, safety, debugging, troubleshooting, abuse prevention, and improvement.
AI providers may process data depending on the configured provider and settings. See Privacy and Data & Processing.