Product overview
ByteGuard Enterprise AI Security Gateway
The control plane, identity model, enforcement flow and deployment coverage in one deck.
Request a copyResources
Learn how ByteGuard governs enterprise AI.
Product material, documentation and answers to the questions security teams ask first.
Resource library
Everything you need to evaluate ByteGuard.
Datasheet
Architecture & requirements
Supported surfaces, endpoint agents, browser extensions and integration points.
Request a copyDocumentation
Deployment & administration
Enrollment, policy and playbook design, agent health and reporting guides for customers.
Contact usFAQ
Short answers to common questions.
Need something more specific? Our team can walk you through it in a demo.
Request a DemoWhere is the policy decision made?
On supported enforcement surfaces the decision is made before the prompt, file or tool call reaches the AI provider. ByteGuard holds the request, evaluates it against policy and then allows, blocks or quarantines it.
What happens with files ByteGuard cannot read?
Unresolved files, such as password-protected or unknown formats, are handled fail-closed and blocked by default.
Which file types are inspected?
PDF, Office documents [DOCX, XLSX, PPTX], structured data [TXT, CSV, JSON] and images [PNG, JPG, TIFF] through OCR.
Which operating systems and browsers are supported?
Endpoint agents run on Windows, macOS and Linux. Browser coverage uses enterprise force-install management for Chrome and Edge and a signed XPI for Firefox.
Does ByteGuard replace our DLP, SWG or SIEM?
No. ByteGuard is a standalone AI security layer that works alongside DLP, EDR, identity, SIEM, SWG, DNS security and firewalls, and adds AI awareness to them.
What does quarantine mean?
A quarantined request is held pending approval instead of being sent to the provider, so sensitive cases receive a human decision.
Can policies differ for the same AI provider?
Yes. The AI Identity Model combines endpoint, OS session, application, surface, provider account and department, so the same provider can fall under different policies.
Does it cover agents and MCP servers?
Yes. ByteGuard provides inventory, usage visibility and policy for MCP servers, tools, skills and project workspace context.
Can ByteGuard manage AI budgets?
Yes. ByteGuard reports AI usage by user, provider, account and department, and can apply quota-based policies such as allow or block when a limit is reached.
Is the list of supported surfaces exhaustive?
No. Coverage is capability-based. The surfaces we list are representative examples, not the full product scope.