Microsoft builds trust-aware controls to address indirect prompt injection
Start here
To reduce the risk of indirect prompt injection, Microsoft is introducing a trust-aware deterministic control layer for M365 Copilot, establishing a trust boundary between tool outputs and the main LLM and classifying content as trusted or untrusted based on deterministic signals (such as emails from outside the organization). Untrusted content is isolated and converted, through a capability-restricted path, into a structured representation that removes malicious content, retains useful information and attaches source labels; the main model receives only trusted original text and the converted structured representation. This layer supplements classifier filtering and system prompt hardening, but Microsoft states that it cannot prevent every indirect prompt injection. In a note published on October 27, 2026, Microsoft added that the protection initially focuses on high-risk content such as external communications, and that tenants need to enable Outlook sender settings and consult the CVE-2026-55145 configuration guide.
Generated by AI from reports · Updated 24 days ago
Progress timeline
- Microsoft Windows Developer BlogMicrosoft builds trust-aware AI controls: handling untrusted content separately
Microsoft is introducing a trust-aware deterministic control layer for M365 Copilot that handles trusted and untrusted content separately to reduce indirect prompt injection risk. Untrusted content is isolated and converted through a restricted path with source labels attached; the main model receives only trusted original text and the converted structured representation. The protection initially focuses on high-risk content such as external communications; tenants need to enable Outlook sender settings and consult the CVE-2026-55145 configuration guide.
- Microsoft Community HubHow Microsoft builds trust-aware controls for M365 Copilot, handling untrusted content differently
Microsoft has added a deterministic enforcement layer at inference time to M365 Copilot, placing a trust boundary between tool outputs and the main LLM, and classifying content as trusted or untrusted based on deterministic signals, such as emails from outside the organization. Untrusted content is converted through an isolated, capability-limited path into a structured representation that removes malicious content, retains useful information, and attaches source markers. This layer supplements classifier filtering and system prompt hardening, but cannot stop every indirect prompt injection.
Related discussion
- Older than 72 hours · Not included in the 72-hour figures
- X@tushigami · うしがみ@Cowork推し![Security] We are introducing Trust-Aware Controls to Microsoft 365 Copilot. External emails and other content are not passed directly to the main model but are processed through a constrained path to reduce the impact of indirect prompt injection. CVE-2026…(opens the original post in a new tab)
12 likes·Oct 16 00:34Collect
- X@RL_landing · RL同学的奇妙旅程Microsoft launches an isolation control layer in Microsoft 365 Copilot to guard against prompt injection. Microsoft has added a security control layer to Microsoft 365 Copilot. External emails are no longer handed directly to the main model but are first converted into plain data in an isolated path. This is to guard against prompt injection atta…(opens the original post in a new tab)
- X@Futureisgood_ · Future-is-good.com : AI lab of EmmanuelVivier.com1⃣2⃣ 🛡 Microsoft locks down Copilot against prompt injection > A deterministic isolation layer for untrusted cont…(opens the original post in a new tab)
1 comments·Oct 16 00:34Collect
- X@emmanuelvivier · Emmanuel Vivier1⃣2⃣ 🛡 Microsoft verrouille Copilot contre l'injection de prompt > Une couche d'isolation déterministe pour bloqu…(opens the original post in a new tab)
1 comments·Oct 16 00:34Collect
- X@tushigami · うしがみ@Cowork推し![Security] We are introducing Trust-Aware Controls to Microsoft 365 Copilot. External emails and other content are not passed directly to the main model but are processed through a constrained path to reduce the risk of indirect prompt injection. Configuration guidance is…(opens the original post in a new tab)
6 likes·Oct 16 00:34Collect
- X@CockerillBill · Bill CockerillMicrosoft 365 Copilot said on October 13, 2026 that it is adding a new enforcement layer that quarantines untrusted con…(opens the original post in a new tab)
1 likes · 1 comments·Oct 16 00:34Collect
- X@sasami_axis · ささみ"Trust-Aware Controls" as a new defense against indirect prompt injection attacks on Microsoft 365 Copilot. Building Trust-Aware Control…(opens the original post in a new tab)