Improper neutralization of special elements used in a command ('command injection') in Microsoft 365 Copilot's Business Chat allows an unauthorized attacker to disclose information over a network.
Severity
Critical
CVSS base score
8.6
Impact
Information Disclosure
Exploitation detected
No known exploitation (MSRC assessment)
Exploitability
Not assessed
Publicly disclosed
No (MSRC assessment)
EPSS · 30-day forecast
0.5%FIRST forecast dated 10 Nov 2026, for the following 30 days · FIRST percentile 45.8% · about EPSS
13 Oct 2026Updated the CVSS User Interaction metric from Required (UI:R) to None (UI:N). The CVSS base score has increased to 8.6. This is an informational change only. No customer action is required.
CVSS 3.1 base metricsRed raises the risk · green lowers it
Attack vector
NetworkAV:N
Attack complexity
LowAC:L
Privileges required
NonePR:N
User interaction
NoneUI:N
Scope
ChangedS:C
Confidentiality
HighC:H
Integrity
NoneI:N
Availability
NoneA:N
Updates that fix it
0 KBs
No KB listed
No KB update is listed in the collected MSRC record. Check the MSRC advisory for product releases and other remedies.
Affected products
Apps
1 affected product
Microsoft 365 Copilot's Business Chat
What changed
Detection template: Unknown: detection-template evidence is incomplete or retained → No Nuclei template listed
Detection template: No Nuclei template listed → Unknown: detection-template evidence is incomplete or retained
Detection template: Unknown: detection-template evidence is incomplete or retained → No Nuclei template listed
Detection template: No Nuclei template listed → Unknown: detection-template evidence is incomplete or retained
Detection template: No Nuclei template listed · Public disclosure: No → No (MSRC assessment) · Exploitation detected: Not reported by MSRC → No known exploitation (MSRC assessment)
Detection template: Unknown: exploit-code evidence is incomplete or retained → No Nuclei template listed · CISA KEV: Unknown: catalog evidence is incomplete or retained → Not listed in the collected catalog
Detection template: No Nuclei template listed → Unknown: exploit-code evidence is incomplete or retained · CISA KEV: Not listed in the collected catalog → Unknown: catalog evidence is incomplete or retained
Publisher revision note: Updated an acknowledgement. This is an informational change only. → Updated the CVSS User Interaction metric from Required (UI:R) to None (UI:N). The CVSS base score has increased to 8.6. This is an informational change only. No customer action is required. · CVSS vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:N/A:N/E:U/RL:O/RC:C → CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:N/A:N/E:U/RL:O/RC:C · CVSS score: 7.4 → 8.6
Page history
Changes are dated when recorded; day views use the last saved version from that UTC day, and citations stay fixed.
14 Oct 2026 · 15:58:54 UTCRevision noteUpdated an acknowledgement. This is an informational change only. → Updated the CVSS User Interaction metric from Required (UI:R) to None (UI:N). The CVSS base score has increased to 8.6. This is an informational change only. No customer action is required.
CISA KEV is observed exploitation evidence. FIRST EPSS is a dated prediction of exploitation in the next 30 days. Missing scores are unknown. This collector requests predictions only for explicit CVE aliases in the public NuGet archive; Microsoft release views retain their existing prediction scope.
No collected risk observation for these identifiers. This does not establish absence of exploitation or a zero probability.
Dated risk observations
Risk history is separate from CNA/OSV content history. Daily score drift is retained but only an upward crossing of 10% triggers the prediction change event.
Immutable source observations
4 recorded revisions
Each citation identifies one complete source statement. Source modification dates and the time this installation observed them are distinct. Latest history includes records historically associated by source identifiers; a removed alias is not a current equivalence claim. Exact citations show that source record's own earlier observations. Failed reads and unchanged polls do not rewrite history.