Overview — Solution Center

01 · Company

Secure every AI interaction

Jan 2020Founded

~500enterprise customers

$185M+capital raised

02 · Platform

One platform: the Secure Enterprise AI Workspace.

AI Security Runtime™

Secure Enterprise AI Workspace

Approved AI, agents, and company knowledge in one secured workspace.

Governed Workspace securing a chat interaction

Browser

coming soon

03 · Solutions

Six solutions run on one security runtime.

AI Security Runtime™

AI Observability

Inventory every AI interaction
Export security evidence to SIEM
Measure adoption and usage

Data Protection

Classify data by meaning
Redact sensitive content in place
Enforce destination policy

Shadow AI Discovery

See unsanctioned AI use
Allow, block, or redirect access
Move users to approved paths

Agent & AI Security

Secure prompts and agent actions
Control tools, APIs, and MCP access
Route exceptions to human review

Cost & Budget Controls

Attribute spend by team and project
Set budgets and usage policy
Route work to approved models

Compliance & Audit

Map policy to requirements
Preserve decision evidence
Report without manual reconstruction

04 · How it works

Inside the AI Security Runtime™

Secure the entry pointSecure Enterprise AI WorkspacePrompts and agent actions enter one governed path.

Control the pathAI Firewall + Gateway

Attach identity
Normalize requests
Control provider access

Understand the interactionSemantic Security

Interpret meaning and intent
Detect sensitive data
Identify threats

Apply the decisionPolicy Enforcement

Allow or redact
Route to human review
Block unsafe actions

Release only what policy allowsModels, tools, and agentsEvery decision produces evidence for security operations and audit.

05 · Semantic Security + Context Graph

Meaning and context drive the decision

Semantic Security

Understands what the interaction means, intends, and could put at risk

Prompts, responses, tool calls, and agent actions
Sensitive data, unsafe objectives, and emerging threats
How intent changes across a sequence of work

Security Context Graph™

Supplies the business context behind every security decision

Who is acting, from which device, and in what role
What data is involved and where it is going
Applicable policy, history, and prior decisions

Combined in real timeAllow, redact, hold for human review, or block

06 · Why semantics matter

Security now has to reason in real time

01

One channel now carries data, instructions, and actions

The content can inform a model, change its behavior, and trigger tools at the same time.

02

The same words can be safe or dangerous

Identity, task, destination, history, and tool access determine the security outcome.

03

There may be no stable signature to match

The same objective can be expressed thousands of ways or emerge across a sequence of steps.

04

Security has to understand the interaction

Semantic Security evaluates meaning, context, threats, and risk before the action proceeds.

Traditional controlsRecognize known indicators and stable patterns

Semantic SecurityReasons about meaning and context in real time

Purpose-trained security AIOnly AI trained for this security task can evaluate meaning, context, threats, and actions at the speed of the interaction.

07 · Enforcement

Policy is enforced before content reaches an outside provider

Our security models are hosted in-house. No outside model provider sees a prompt unless the AI Security Runtime™ clears it.

Before the boundaryPrompt or agent actionOutside providers have not seen the content.

First Recon security boundary

AnalyzeMeaning, intent, data, threats

ContextualizeIdentity, device, destination, policy

EnforceAllow, redact, review, or block

Our security models run inside First Recon infrastructure.

After the boundaryCleared content onlyThe outside provider receives only what policy allows.

08 · Deployment

Deploy where AI work happens

Secure Enterprise AI Workspace

Where it fits: Employees need approved AI now
Best for: Chat, agents, knowledge, and collaboration
Rollout: Browser and desktop access in hours

FromScattered AI toolsToOne secured workspace

Live in hoursBrowser and desktop access in hours, with baseline policies applied from the start.

09 · Time to value

Security value in hours. Deeper control over time.

Hours

Start with 300+ policies

Policies are automatically selected and applied for your industry, sector, region, and compliance requirements.

Provision workspace
Approve providers
Invite users

Days

Bring your policy with you

AI-assisted import turns existing data and security policies into First Recon policy drafts for your team to review.

Upload policy
Connect identity
Review and publish

Weeks

Broaden coverage

Connect company knowledge and broaden coverage across more teams and workflows.

Add knowledge sources
Onboard more teams
Connect security tools

Ongoing

Improve the program

Tune controls using adoption, security decisions, spend, and threat evidence.

Refine policy
Expand enablement
Measure results

Deployment and ongoing technical support are included throughout.First Recon works with your team from initial setup through program expansion.

10 · Secure Enterprise AI Workspace

Chat

Every major model in one governed chat, secured before content leaves the workspace.

Model choice without security sprawl

Governed Workspace Chat with sensitive data protection

11 · Secure Enterprise AI Workspace

Cowork Agent

Run multi-step work with reusable agents inside the same governed workspace.

Choose the right model for the work

Governed Workspace Cowork Agent active session

12 · Secure Enterprise AI Workspace

Agent Hub

Discover, configure, and oversee governed cloud agents and teams.

Govern agents across model providers

Agent Hub workspace with an active governed agent and work queue

13 · Secure Enterprise AI Workspace

Knowledge

Vector search and RAG retrieve the right approved knowledge at scale for people and agents while preserving source permissions.

Ground every supported model in approved knowledge

Governed Workspace Knowledge across approved company sources

14 · Secure Enterprise AI Workspace

Team Feed

Turn useful AI work into reusable examples that improve team enablement and productivity.

Share what works across every approved model

Governed Workspace Pulse feed for enablement and productivity

15 · Security operations

From sprawl to one platform.

AI Gateway

Cloudflare

Portkey

Kong

AWS

AI Firewall

Cloudflare

Prisma AIRS

Fortinet

Snyk

Secure Enterprise AI Workspace

Glean

Dust

Team-GPT

Writer

AI Observability

Datadog

Arize

Grafana

W&B

Compliance & Audit

Credo AI

Holistic AI

IBM

ServiceNow

Cost & Budget Controls

Finout

CloudZero

Apptio

CloudHealth

AI Security Runtime™

Six categories. One secured platform.

One enforcement layer, policy system, and decision trail.

16 · Enterprise architecture

Secure AI interactions at the control point

Managed appMany ungoverned browser and desktop tools

Secure Enterprise AI Workspace

One approved workspace for chat, agents, and company knowledge.

Connected trafficAPIs, MCPs, and agent actions

AI Security Runtime™

Apply policy before the destination and record the decision.

17 · Trial

A 30-day trial, live in hours

30 daysUp to 50 users$500 AI creditsFull security runtime

Hour 1Provision

Workspace, admins, and baseline policies

Day 1Invite

Trial admins add users and approved providers

Days 1–30Work

Use normal workflows and company knowledge

WeeklyLearn

Review adoption, security decisions, and coverage

Day 30Decide

Plan the broader deployment

Self-service onboardingGuided setup is available.

18 · Support

Deployment and ongoing technical support are included

Deploy

Launch with the team

Architecture, setup, policy configuration, and rollout support.

Operate

Get help in context

First Recon support agent, human ticketing, and in-platform support.

Scale

Stay connected

Dedicated Slack or Teams channels with a global 24-hour response SLA.

24-hourresponse SLA

Global support · dedicated channels

19 · Enterprise AI Training & Enablement

Build AI capability across the enterprise

First Recon Academy learner dashboard with role-based tracks and course progress

First Recon Academy supports business professionals, executives, and technical teams, not only platform users. Content is updated regularly.

FoundationalAlways freePractitionerProfessionalMaster

6 tracks34 courses187 lessons2 certification exams

AI & Agent Usage for Knowledge Workers 6 courses
Agentic Coding & SDLC 6
Agentic Engineering & Agent Management 6
Cybersecurity for AI 6
First Recon Platform 7 + certification
Identity for Agents 3 + certification

20 · The transformation

From AI sprawl to one secured platform

Before

Employees and security teams manage separate AI tools

Each AI app and security product adds another policy, data path, evidence store, and operating layer.

Employee AI tools

AI security tools

One approved AI experience

The Secure Enterprise AI Workspace, backed by AI Security Runtime™.

One security control layer
One policy across every AI path
One decision record

Start with a 30-day trial · up to 50 users

Secure every AI interaction

Solution Center

Secure Every AI Interaction

Secure every AI interaction

One platform: the Secure Enterprise AI Workspace.

Secure Enterprise AI Workspace

Six solutions run on one security runtime.

AI Observability

Data Protection

Shadow AI Discovery

Agent & AI Security

Cost & Budget Controls

Compliance & Audit

Inside the AI Security Runtime™

Meaning and context drive the decision

Understands what the interaction means, intends, and could put at risk

Supplies the business context behind every security decision

Security now has to reason in real time

One channel now carries data, instructions, and actions

The same words can be safe or dangerous

There may be no stable signature to match

Security has to understand the interaction

Policy is enforced before content reaches an outside provider

Deploy where AI work happens

Secure Enterprise AI Workspace

Security value in hours. Deeper control over time.

Start with 300+ policies

Bring your policy with you

Broaden coverage

Improve the program

Chat

Cowork Agent

Agent Hub

Knowledge

Team Feed

From sprawl to one platform.

AI Security Runtime™

Secure AI interactions at the control point

Secure Enterprise AI Workspace

AI Security Runtime™

A 30-day trial, live in hours

Deployment and ongoing technical support are included

Launch with the team

Get help in context

Stay connected

Build AI capability across the enterprise

From AI sprawl to one secured platform

Employees and security teams manage separate AI tools

One approved AI experience

Talk to the team