Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

Microsoft Mechanics Podcast

Microsoft Entra Suite hands-on tour of identity and network access protections

Jul 15, 20269 minVideo
ShareXLinkedInFacebookWhatsAppEmail

Unify identity and network access controls. Enforce least privilege access across every app and resource. Wire lifecycle workflows directly to your HR system to strip stale permissions on role changes, gate sensitive data behind biometric step-up verification, and replace your VPN with per-app, identity-scoped access that revokes tokens the moment risk spikes. Secure AI usage at every layer. Block confidential data from reaching public AI tools and stop adversarial prompt injections before your agents process them. John Damon, Microsoft Entra Suite Senior Product Manager, shares how to lock down identity, network access, and AI usage from a single control plane. ► QUICK LINKS: 00:00 - Identity and network controls 00:45 - Lifecycle Workflows 01:28 - Verified ID with Face Check 02:15 - Request access for direct reports 03:17 - Global Secure Access + Token Revocation 04:32 - Secure AI usage 06:20 - Network DLP / ChatGPT Block 07:12 - Prompt Injection Protection 08:31 - Wrap up ► Link References Check out our related deep dives at https://aka.ms/EntraSuitePlaylist For more information, go to https://aka.ms/EntraSuite ► Unfamiliar with Microsoft Mechanics? As Microsoft's official video series for IT, you can watch and share valuable content and demos of current and upcoming tech from the people who build it at Microsoft. • Subscribe to our YouTube: https://www.youtube.com/c/MicrosoftMechanicsSeries • Talk with other IT Pros, join us on the Microsoft Tech Community: https://techcommunity.microsoft.com/t5/microsoft-mechanics-blog/bg-p/MicrosoftMechanicsBlog • Watch or listen from anywhere, subscribe to our podcast: https://microsoftmechanics.libsyn.com/podcast ► Keep getting this insider knowledge, join us on social: • Follow us on Twitter: https://twitter.com/MSFTMechanics • Share knowledge on LinkedIn: https://www.linkedin.com/company/microsoft-mechanics/ • Enjoy us on Instagram: https://www.instagram.com/msftmechanics/ • Loosen up with us on TikTok: https://www.tiktok.com/@msftmechanics

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show.

0:02With AI, where action happens at machine speed and where access can be granted and inherited instantly, identity and network access has never been more important. Securing AI starts with securing people, and for that, your identity and network controls need to come together to close gaps that attackers can exploit, and that's where Microsoft Entra Suite comes in. It combines best-in-class capabilities into a single solution to help you enforce least-privilege access to make sure users have access to what they need, and only as long as necessary.

0:35Apply unified access controls to any app and resource, and secure access to AI by discovering AI apps and agents, assessing risk, and enforcing policy. Let's bring this to life by following a user, Violet Martinez, throughout her day, I'll start by showing how she gets access to exactly what she needs, with least-privilege access. In this scenario, our user has changed job roles. Her role change was signaled overnight by an HR system, Workday, and her permissions need to be adjusted for her new position in the IoT department.

1:05This is where a lifecycle workflow we set up in advance in Entra Suite comes into play. We've created a "Mover" workflow that automatically adjusts access when a user role changes, and I can click to see how it's defined. Importantly, stale entitlements from her previous role are automatically removed, and a new access package bundling the right apps and permissions for her new IoT product marketing role is automatically assigned. Now let's switch to showing you our user Violet's experience. When she signs in with a passkey for her new role, every app and permission

1:36from the baseline package is already set up, and she can request access to specific resources that she might be missing using the My Access page at myaccess.microsoft.com. For example, in order to start her work in competitive analysis, she needs access to Zava's on-prem IoT Pricing Dashboard. So she makes the request. Because the package grants access to highly confidential on-prem pricing data, the workflow requires Verified ID step-up with Face Check. This requires Microsoft Authenticator, which prompts for Face Check,

2:07a verification process to match her real-time selfie to her government-issued ID on file, and once verified, she's given access to the dashboard. Now, as a manager, you can request access for your direct reports to make it easier for users to onboard with the access they need. In fact, let me show you the perspective from Violet's manager. In the same My Access page, I can see the Pricing Dashboard access package we just saw, along with another access package that I can assign to my team. In this case, I want to extend the access to the Zava Assistant AI agent used by the team,

2:38and can initiate an access package request for the agent on Violet's behalf. From the dropdown, I see the directs on my team, I'll choose Violet. Now I'll choose a start date followed by an end date, and along with those, I'll type in the business justification, which was set up by IT as mandatory properties for this access package. And because in this instance, as the manager, I am both the requester and approver, Entra ID Governance both provisions and grants just-in-time and time-bound access. Least-privilege access is enforced automatically,

3:10with stale permissions removed and updated baselines assigned, and sensitive access controlled using step-up verification. From here, let's move on to applying real-time context-aware access controls to apps and resources as our user goes about her day. She starts in the Edge browser and navigates to local IP to open on-prem Pricing Dashboard. We can see that she has the Global Secure Access client installed, so she doesn't need to use a VPN. To access internal resources, our IT policy requires her to sign in using her work account with passkey,

3:43which uses Conditional Access to evaluate her risk and session context before allowing access to on-prem Pricing Dashboard. Importantly, Global Secure Access permissions are granted per app and scoped to identity, with no inbound firewall ports and no public IPs exposed. And by the way, Global Secure Access will also work with other on-prem apps and Active Directory that do not natively support modern authentication. The good news is, even if her device is compromised because of a hardware-based token theft and a token replay attack,

4:14with dynamic policies in place, once the user risk is flagged as elevated, token access can be automatically revoked. This forces self-remediation for any user account with elevated risk. That way, privilege never accumulates and trust is continuously reevaluated without standing privilege to stop identity attacks. Next, we already saw that our user is expected to use AI as part of her job, but how does Entra Suite help with securing AI usage? Let's take a look. When our user leaves the office, she uses her personal laptop to work on an FY27 presentation

4:46from her team's SharePoint site. She's in Microsoft Edge, and using her personal browser profile. A SharePoint site is bookmarked, and she tries to access it. In order to get to the protected location on her laptop, she signs in with her work account, and Conditional Access requires an app protection policy, which then prompts her to switch the Edge profile to her work account. When she does, the device is now registered and the app protection policy is delivered to the browser. This automatically applies security settings and policies to the work browser profile. It enables explicit forward proxy settings

5:18and TLS inspection to add visibility into encrypted traffic, so that the company's data protection policies can work with it. This includes Microsoft Edge data loss protection controls, as well as Microsoft Entra Internet Access, Secure AI and Web gateway policies. That way, she can access her work documents securely, like you're seeing here with the internal FY27 planning presentation. That said, when using her work profile in the browser, if she opens a new tab and she tries to access a social media site, we can see access is blocked

5:48based on the company policy. But once she switches back to her personal profile, access to Facebook works as expected. Policies are pushed automatically, even though she's not using a managed device. In fact, Entra Suite inspects and controls access to data. As she interacts with any website, her traffic routes through Entra Internet Access as a forward proxy. So every egress path is inspected in real time against the Purview sensitivity label applied at the source. Same label, same policy, whether she's on a corporate laptop

6:19or a personal device. This time, our user returns to work with her managed work laptop, and she wants to analyze the data from the Pricing Dashboard using ChatGPT. She has a confidential internal product pricing schedule opened as a PDF and selects everything in the document, and she copies everything on the page. Then she moves over to ChatGPT and pastes the pricing info from her clipboard. Here, Secure Web and AI Gateway in Entra Suite blocks the upload because there's an organization-wide policy that prohibits sharing confidential data

6:49through a public AI tool. This adds an important layer of protection because a standalone web gateway or CASB can only see traffic, but not the data classification. This is made possible because network DLP parsed the chat text to spot sensitive information before ChatGPT saw the payload, and network DLP also will block the sharing of sensitive files over non-Microsoft email services like Gmail. Next, let's look at the secure AI usage when the user leverages her company-sanctioned AI tool. Here's the Zava Assistant that her manager approved.

7:20She starts opening a competitor blog and copies everything into her clipboard. Then she invokes the agent and starts to interact by pasting in the blog for summarization, but there's a catch. Hidden inside the text is a human-invisible instruction telling the model to leak her query history. Here, prompt injection protection inspects the outbound prompt against Microsoft's adversarial pattern model, matches the known injection class, and blocks it before it's processed. Additionally, new web filtering rules let you block agents from conducting risky operations on specified resources,

7:52by creating policies for browser-based and local apps communicating over web protocols. This time, our user remembers that she's left a sensitive Zava partner memo in the shared Dropbox location. So, using her locally installed Claude app, she requests another in-house developed Zava agent to remove that file from Dropbox. Immediately, she can see that this action was prevented based on the network policy from her company. And moving to the admin experience, these rules can differ whether it's a user or agent session performing the operation. They can be configured to assess multiple HTTP methods,

8:25including POST, PATCH, PUT, and DELETE operations scoped to specific URLs or FQDNs. Those were just a few examples of Microsoft Entra Suite and its unified approach to access. With Microsoft 365 E7, you can get Entra Suite protections and apply them to AI agents with Agent 365. Check out our related deep dives at aka.ms/EntraSuitePlaylist, and to learn more, go to aka.ms/EntraSuite. Subscribe to Microsoft Mechanics

8:55for the latest tech updates, and thanks for watching.

Open the transcript file

Hosts and Guests

Nobody is credited on this episode yet.

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. Introducing Autopilot, your personal and proactive agent

    Nov 20, 202610 minVideo

    Put an AI teammate to work around the clock with Autopilot in Microsoft Copilot, a persistent, proactive and personal agent that runs in the cloud and keeps working after you power off your laptop. Describe your goals...

    Credit the people on this episode

  2. Build apps by describing what you want | Copilot Code

    Nov 19, 20268 minVideo

    Turn your app idea into a working, interactive web app with Copilot Code. Describe what you want in plain language, ground it in your work data with Work IQ, build what-if modeling from your own business logic, and...

    Credit the people on this episode

  3. Manage AI Cost + ROI with Copilot and Agent 365

    Nov 19, 202610 minVideo

    Take control of AI spend across Microsoft Copilot, Cowork, and agents inside and outside of Microsoft. Track your own Copilot Credit usage task by task, pin down your top cost drivers by service and model provider in...

    Credit the people on this episode

  4. Azure SQL Database Hyperscale - Now for any size workload

    Nov 18, 202614 minVideo

    Start small and scale up to your most demanding workloads on Azure SQL Database Hyperscale. Run serverless compute that scales with demand and auto-pauses, so you stop paying for idle compute. Pool databases across...

    Credit the people on this episode

  5. New Microsoft Copilot Harness | More power, token efficiency & control for agents in Foundry

    Nov 10, 202615 minVideo

    Build agents you can trust with your company's data, apps, and resources using Microsoft Foundry. Start with a fully managed prompt agent and move it onto the Microsoft Copilot harness so it plans, loops, calls...

    Credit the people on this episode