Assess before acting
Discover the authorised environment and examine its security posture. Make incomplete checks and detection gaps visible.
AI Defence / The Remedy security programme
AI security audit. Continuous defence.
See the risk. Understand the recommendation. Control the response.
AI Defence is designed to assess your networks, servers and applications, explain what matters, and verify authorised security changes before establishing a baseline for continuous monitoring.
IN DEVELOPMENTExplore the specified scope. See capability availability.
DEFENCE / OPERATING MODELREAD-ONLY FIRST
Networks
Servers
Applications
AI DEFENCEEvidence → Context → Risk
Clear report
Your authority
Verified outcome
Change boundary
No remediation before authorisation.
APPROVAL REQUIREDNo remediation before authorisation.
ENGINEERED WITHIN ALTARIRead-only assessmentCustomer-controlled changesEvidence-backed verification
The whole environment, in context
A security finding is only the beginning.
AI Defence connects what is exposed, what is vulnerable and what has authority. The result is designed to be a decision you can act on, with the evidence to support it.
Explain the real risk
Translate technical evidence into a clear report: what matters, why it matters, and what a change would affect.
Prove the outcome
Apply only authorised remediation. Retest the security condition and the services that need to keep working.
25Defined assessment & protection areas
Read-onlyFirst-entry assessment by default
Your callAuthorise all, selected or no changes
VerifiedThe required outcome of a correction
Security coverage
Follow the risk across the boundaries.
The specified scope reaches beyond individual hosts to the identities, data and delivery systems they depend on.
Network security
Network paths, exposed services, segmentation and the boundaries between systems.
Explore this security area ↗Server security
OS posture, patch levels, services, permissions and suspicious runtime changes.
Explore this security area ↗Cloud & container security
Workload privilege, Docker exposure, image provenance and cloud access.
Explore this security area ↗Identity & secrets
Accounts, machine identities, SSH keys, excessive permissions and exposed credentials.
Explore this security area ↗Application & API security
Authentication, sessions, API controls, TLS and application exposure.
Explore this security area ↗Database & data security
Data access, encryption, tenant boundaries, backups and audit evidence.
Explore this security area ↗Also explore contextual vulnerability management and software supply chain security.
Risk correlation
Same vulnerability.
Different consequences.
An exposed, privileged service can create a different risk from the same software in an isolated test environment.
AI Defence is designed to weigh exploitability, exposure, business importance, privilege and compensating controls together.
How risk is prioritised ↗ILLUSTRATIVE RISK SCENARIO
Look at the combination.
Internet exposureVulnerable serviceKnown exploit evidenceElevated privilege
Prioritise investigation.
Explain the evidence, identify the affected boundary and recommend a controlled response.
A risk example, not a live finding or claim of confirmed compromise.
A governed lifecycle
From first inspection to a defensible baseline.
01
Discover & audit
Inventory the environment. Inspect controls. Record evidence and coverage gaps.
02
Explain & recommend
Correlate risk. Produce the report. Explain impact and priority.
03
Authorise & verify
Agree the action. Prepare recovery. Apply the change and retest.
04
Baseline & monitor
Record the accepted state. Detect new exposure, vulnerabilities and drift.
Authority stays with you
The ability to detect a problem is not permission to change your environment.
A deliberate approval boundary.
The first assessment is observational. Recommendations explain possible downtime, dependencies and the order of work. You can authorise everything proposed, selected changes, individual findings or nothing.
Material changes need current-state checks, impact analysis, rollback preparation and protected administrator access.
Read the remediation model ↗One engineering family. Distinct responsibilities.
Security, repair and delivery.
AI Defence
Security posture, authorised hardening and continuous threat and vulnerability monitoring.
Programme status ↗Remedy
Software defect diagnosis, source-code repair and verification. From bug report to verified fix.
Remedy AI Bug Fixer ↗Nexus
The governed autonomous software engineering control plane for delivery around AI workers.
Nexus software engineering ↗Start with understanding
What does your environment need to defend?
Discuss the assessment scope, current capabilities and the control you need with Altari Systems.