Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Improving posture to get ready for AI means establishing a defensible device, application, data, software, and infrastructure baseline before AI systems and agents inherit the environment. The work reduces reachable attack surface, closes common control gaps, and prioritizes remediation where it reduces the most business risk.
This guidance provides detailed implementation continuity for work described in the Secure your foundations AI security transformation theme. Use the Microsoft Security Adoption Framework (SAF) to coordinate this work across business scenarios, security disciplines, roles, technologies, and technical solutions.
Why this work matters
AI gives attackers new ways to discover, analyze, and chain vulnerabilities at greater speed. Many of the weaknesses they exploit remain familiar:
- unmanaged or noncompliant devices;
- organizational data leaving managed applications;
- unnecessary internet-facing exposure and attack paths to critical assets;
- unpatched software and operating systems;
- vulnerable open-source dependencies and software supply chains;
- weaknesses in application and infrastructure code; and
- unmanaged identities, credentials, and secrets.
AI adoption raises the stakes because AI systems and agents can inherit the posture of the devices, identities, services, data, software, and infrastructure they can reach. Strengthening these foundations reduces risk before AI increases scale and connectivity.
Prioritize posture improvements
Use Secure Now and Microsoft Security Exposure Management to review current recommendations, prioritize work, and track progress. Connect the implementation work to SAF so priorities, ownership, and measures are sustained after an individual initiative is complete.
| Initiative | Durable outcome | Primary SAF connection | Implementation resource |
|---|---|---|---|
| Manage organizational devices | Devices that access organizational resources are known, managed, monitored, and assigned to accountable owners. | Security posture management, endpoints | Manage devices with Intune |
| Implement device compliance requirements | Access decisions include device health and policy compliance instead of relying on identity alone. | Access and identities, endpoints | Set up compliance policies and require device compliance |
| Protect organizational data in managed applications | Work data remains protected in approved applications, including on devices the organization does not manage. | Data security, endpoints | Protect apps and implement data loss prevention |
| Reduce internet-facing exposure | Unnecessary exposure and reachable attack paths to critical assets are removed or controlled. | Security posture management, infrastructure, networks | Microsoft Security Exposure Management |
| Improve patching and vulnerability management | Remediation is prioritized by asset criticality, reachability, exploitability, and business impact. | Security posture management, infrastructure security | Secure Now |
| Secure the software supply chain | Dependencies, build systems, artifacts, and publishing processes use approved, observable security controls. | Development security, development security implementation patterns | Microsoft Security Development Lifecycle and protect the software supply chain |
| Scan and remediate code | Security weaknesses are found and fixed through repeatable engineering workflows before and after deployment. | Development security, applications | Secure source code with GitHub Advanced Security |
Coordinate the work through SAF
Treat the initiatives as connected parts of the security program:
- Use Strengthen posture and compliance to define the overall business outcome and measures.
- Use Adopt and secure AI to connect foundation work to AI adoption and AI-specific risk.
- Use Security posture management to maintain inventory, exposure analysis, prioritization, remediation, and verification.
- Use Security roles and responsibilities to assign accountable owners and cross-team decision rights.
- Use technologies and technical solutions to reach canonical implementation guidance.
- Use the AI security workshop to assess implementation areas and create a roadmap.
Outcomes and measures
Expected outcomes include:
- managed and compliant devices use an approved security baseline;
- data remains protected in managed applications and approved repositories;
- attack paths from internet-facing systems to critical assets trend down;
- vulnerability and patch work is prioritized according to business risk;
- software dependencies, pipelines, code, and artifacts use repeatable security controls; and
- AI systems and agents inherit a governed, measured foundation rather than unmanaged exposure.
Consider measures such as:
- percentage of critical assets with a named accountable owner;
- percentage of corporate devices enrolled in management and compliant with policy;
- percentage of applicable users covered by application protection policies;
- number of reachable attack paths from internet-facing assets to critical assets;
- patch and vulnerability remediation performance by asset criticality and service-level objective;
- percentage of production builds that use approved pipelines, code scanning, dependency controls, and software bills of materials; and
- Secure Now or exposure-management initiative progress compared with the organization's target and baseline.
Next steps
- Review Secure your foundations to connect this implementation work to the broader AI security outcome.
- Review AI security transformation to understand how current AI security themes translate into SAF execution.
- Continue with Build a strong security posture for AI and prepare your environment for AI security.