Junus AI Shield

Junus AI Shield is the browser companion to the Junus AI Shield Control governance plane. It helps organizations see and govern how employees use AI tools (ChatGPT, Claude, Gemini, Copilot, Perplexity and similar) directly in the browser, and enforces your organization's AI-usage policies on-device — warn, block, require justification, or redact sensitive data before it reaches an AI tool. HOW IT WORKS • Detects when you open a recognized AI tool and records it as a usage event for your organization's Shield Control workspace (an inventory of "who uses what"). • Enforces the policies your administrators configure: a prompt can be allowed, warned, blocked, or held for justification, and sensitive values can be redacted on-device. • A side panel shows recent activity, the active policies, and the status of the current tab. WHAT IS COLLECTED — AND WHAT IS NOT (prominent disclosure) • Collection happens ONLY after you give affirmative consent. On first run the extension shows a consent prompt; nothing is collected until you accept. • You can opt out at any time on the extension's options page, which turns all collection off. • When enabled, the extension records: the NAMES of the AI services you visit, the TYPE of action (e.g. "a prompt was submitted") and its OUTCOME, and — when on-device detection is on — CATEGORY LABELS of sensitive data (e.g. "pii", "financial"). It is sent only to your organization's own Shield Control backend. • It NEVER collects or transmits your prompts, pasted text, file contents, or the sensitive values themselves. Page content is read locally, on your device, for detection only. • Managed (org-deployed) installs: an administrator may enforce collection on behalf of the organization; in that case the opt-out is locked by policy. Privacy policy: https://www.junus.ai/privacy/shield-web Junus AI Shield is for organizations that have deployed Junus AI Shield Control. You sign in with your Shield Control account; the extension does not create accounts.