# 3AM ransomware affiliate using AI loop against UK schools

- ID: bitdefender-3am-ransomware-ai-2026
- Reported: 2026-11-04
- Researcher: Marius Baciu, Adrian Schipor, Martin Zugec
- Organization: Bitdefender
- AI role: attacker_uses_ai_as_operator (AI as operator)
- Confidence in AI involvement: high
- Research / borderline: false
- Last reviewed: 2026-11-05
- Entry page: https://openattacks.ai/attacks/bitdefender-3am-ransomware-ai-2026/
- Entry JSON: https://openattacks.ai/api/attacks/bitdefender-3am-ransomware-ai-2026.json
- Browse title: 3AM ransomware affiliate using AI loop against UK schools
- Editorial summary: 3AM ransomware affiliate used AI loop against UK schools showing model-style command errors including Bash escape sequences in Windows commands and malformed for /f syntax passing K" to comsvcs. Numbered…
- Evidence class (catalog annotation): Not classified
- Source-stated target environments: Not specified
- Stage hunts and conditional defenses: https://openattacks.ai/api/defenses/bitdefender-3am-ransomware-ai-2026.json

## Technique
3AM ransomware affiliate used AI loop against UK schools showing model-style command errors including Bash escape sequences in Windows commands and malformed for /f syntax passing K" to comsvcs. Numbered retry scripts (av\_remove2, av\_remove3) indicate automated iteration. Probe files da\_proof.txt and psexec-ok suggest autonomous validation. Backdoor admins library and svc\_bkpmon created. Staging registry path HKLM\\SOFTWARE\\FIN001. Attack demonstrates AI-driven operations with observable execution errors and iteration patterns.

## Stack relevance (editorial)
Assessment: not\_assessed. No reviewed applicability annotation is available. Product mentions do not establish victim conditions.
A product mention alone is not a victim-side applicability mapping. Selection does not verify affected versions, permissions or compromise.

## Initial access
unknown

## AI tooling and models
- AI-driven command loop (model unspecified)

## Other tooling
- 3AM ransomware
- Numbered retry scripts (av\_remove2, av\_remove3)
- Probe files (da\_proof.txt, psexec-ok)
- Backdoor accounts (library, svc\_bkpmon)
- Registry staging (HKLM\\SOFTWARE\\FIN001)
- PsExec
- comsvcs.dll

## Victims and sectors
UK schools

## Target geography
- United Kingdom
United Kingdom reflects targeted school systems

## Attribution
3AM ransomware affiliate

## Scale and metrics
Multiple UK schools targeted; exact count unknown

## Attack flow
Catalog reconstruction of the supplied source account. Unknowns and research caveats are preserved.
### initial-access: Initial access to UK schools [unknown]
Initial access vector not disclosed in source
Evidence: https://www.bitdefender.com/en-us/blog/businessinsights/3am-ransomware-hybrid-ai-attack

### ai-commands: AI loop generates attack commands [ai]
Model generates commands with observable errors including Bash escapes in Windows commands and broken for /f syntax
Evidence: https://www.bitdefender.com/en-us/blog/businessinsights/3am-ransomware-hybrid-ai-attack

### iteration: Automated retry with numbered scripts [ai]
Numbered scripts av\_remove2, av\_remove3 show iteration; probe files da\_proof.txt, psexec-ok validate steps
Evidence: https://www.bitdefender.com/en-us/blog/businessinsights/3am-ransomware-hybrid-ai-attack

### persistence: Backdoor accounts and registry staging [execution]
Backdoor admins library and svc\_bkpmon created; staging in HKLM\\SOFTWARE\\FIN001
Evidence: https://www.bitdefender.com/en-us/blog/businessinsights/3am-ransomware-hybrid-ai-attack

### ransomware: 3AM ransomware deployed [impact]
3AM ransomware payload deployed after preparation
Evidence: https://www.bitdefender.com/en-us/blog/businessinsights/3am-ransomware-hybrid-ai-attack

Connections:
- initial-access -> ai-commands
- ai-commands -> iteration
- iteration -> persistence
- persistence -> ransomware

## Stage defense and hunting recommendations
Editorial recommendations, not validated coverage. Source-controlled query starters are drafts. Custom-schema examples require telemetry collection and field mapping; absence of a mapping means not assessed. MagicSword policy opportunities require managed Windows scope and enforcement.
### initial-access: No source-backed runtime hunt yet
- Visibility: Not established
- Query origin: catalog_authored; status: not_applicable; log scope: gap
- Hunt hypothesis: The supplied account does not establish a deployed victim-side behavior for this stage.
- Required telemetry: not assessed
- Investigate: Obtain a source-backed procedure and identify which party could observe it before drafting a detection.
- Limitation: Marketing claims, unknown steps and unconfirmed deployments are not detection coverage. No rule is generated for this stage.
- Qualification: The source does not disclose the initial-access procedure, so no runtime query is displayed for this stage.
- No direct MagicSword prevention claim is recorded for this stage.

### ai-commands: Suspicious Windows script execution
- Visibility: Defender-side logs where collected
- Query origin: catalog_authored; status: draft; log scope: native
- SIGMA query download: /api/detections/windows-shell.sigma.yml
- KQL query download: /api/detections/windows-shell.kql
- SPL query download: /api/detections/windows-shell.spl
- Hunt hypothesis: A script host or shell runs encoded content, a remote URL, or inline script syntax, then downloads or launches a payload.
- Required telemetry: Windows process creation: Sysmon Event 1 or Defender DeviceProcessEvents; PowerShell script-block logging when enabled; Download / network and installer events
- Investigate: Follow the parent and child process tree and recover the script or URL from logs without executing it.
- Investigate: Correlate the payload's signer/hash, persistence, and network activity; baseline approved automation.
- Limitation: Encoded commands and remote administration can be legitimate. This does not cover every script host, in-process execution, or non-Windows branch.
- Qualification: Apply the process hunt to the generated Windows commands and their parent process; syntax errors alone are not malicious.
- No direct MagicSword prevention claim is recorded for this stage.

### iteration: Suspicious Windows script execution
- Visibility: Defender-side logs where collected
- Query origin: catalog_authored; status: draft; log scope: native
- SIGMA query download: /api/detections/windows-shell.sigma.yml
- KQL query download: /api/detections/windows-shell.kql
- SPL query download: /api/detections/windows-shell.spl
- Hunt hypothesis: A script host or shell runs encoded content, a remote URL, or inline script syntax, then downloads or launches a payload.
- Required telemetry: Windows process creation: Sysmon Event 1 or Defender DeviceProcessEvents; PowerShell script-block logging when enabled; Download / network and installer events
- Investigate: Follow the parent and child process tree and recover the script or URL from logs without executing it.
- Investigate: Correlate the payload's signer/hash, persistence, and network activity; baseline approved automation.
- Limitation: Encoded commands and remote administration can be legitimate. This does not cover every script host, in-process execution, or non-Windows branch.
- No direct MagicSword prevention claim is recorded for this stage.

### persistence: New persistence → recurring execution
- Visibility: Defender-side logs where collected
- Query origin: catalog_authored; status: draft; log scope: custom_schema
- SIGMA query download: /api/detections/persistence.sigma.yml
- KQL query download: /api/detections/persistence.kql
- SPL query download: /api/detections/persistence.spl
- Hunt hypothesis: A task, Run key, cron entry, shell profile, or startup file is changed and later launches the same unexpected program.
- Required telemetry: Platform-specific registry, scheduled-task, cron and startup-file changes; Process creation and file provenance; User / administrative change records
- Investigate: Diff the persistence artifact and identify the creating process / principal.
- Investigate: Correlate the subsequent launch and network behavior; baseline deployment and management agents.
- Limitation: Windows and Unix persistence require different collectors. A configuration change without a subsequent launch is not proof that persistence ran.
- No direct MagicSword prevention claim is recorded for this stage.

### ransomware: Collection / encryption behavior
- Visibility: Defender-side logs where collected
- Query origin: catalog_authored; status: draft; log scope: custom_schema
- SIGMA query download: /api/detections/impact.sigma.yml
- KQL query download: /api/detections/impact.kql
- SPL query download: /api/detections/impact.spl
- Hunt hypothesis: A process rapidly enumerates or modifies sensitive files and is followed by encryption, export or a ransom-related artifact.
- Required telemetry: File-operation rates and process attribution; Ransomware / DLP behavior alerts; Canary files and recovery / backup audit
- Investigate: Correlate file bursts, extension changes and outbound transfer to the same process and identity.
- Investigate: Validate against bulk backup / archiving activity and preserve the research-only qualification of a PoC.
- Limitation: A file-write burst is not proof of ransomware; in-process interpreters may not spawn a separate executable.
- No direct MagicSword prevention claim is recorded for this stage.


## MITRE ATLAS
Mappings are catalog annotations based on described behavior, not assignments or endorsements by MITRE.
- AML.T0016.002: Generative AI — AI loop autonomously executed attack commands with observable model errors and iteration patterns
  Technique: https://atlas.mitre.org/techniques/AML.T0016.002
  Evidence: https://www.bitdefender.com/en-us/blog/businessinsights/3am-ransomware-hybrid-ai-attack

## Defensive notes
Detect model-style command syntax errors; monitor for numbered retry scripts; alert on probe files in unusual locations; review backdoor account creation; monitor registry key HKLM\\SOFTWARE\\FIN001; detect PsExec lateral movement; hunt for comsvcs.dll credential dumping

## Sources
- https://www.bitdefender.com/en-us/blog/businessinsights/3am-ransomware-hybrid-ai-attack — 3AM Ransomware Hybrid AI Attack

## Source notes
Relevant detection rules: - Sigma: Suspicious LSASS Process Access (comsvcs.dll dump pattern) (https://github.com/SigmaHQ/sigma/blob/master/rules/windows/process\_creation/proc\_creation\_win\_sysinternals\_procdump\_lsass.yml) - Sigma: Process Memory Dump Via Comsvcs.DLL (https://github.com/SigmaHQ/sigma/blob/master/rules/windows/process\_creation/proc\_creation\_win\_rundll32\_process\_dump\_via\_comsvcs.yml) - Monitor for registry key creation under HKLM\\SOFTWARE\\FIN001 - Detect backdoor account creation (library, svc\_bkpmon) - Alert on numbered retry scripts (av\_remove2, av\_remove3 patterns) - Monitor for probe files (da\_proof.txt, psexec-ok) in unusual locations - Detect command syntax errors indicating automated generation (Bash escapes in Windows commands) - LOLBAS references: comsvcs.dll (https://lolbas-project.github.io/lolbas/Libraries/Comsvcs/), PsExec (https://lolbas-project.github.io/lolbas/OtherMSBinaries/PsExec/) - Note: AI-generated command errors and iteration patterns require custom detection logic analyzing command line syntax anomalies. 

## Indicators (untrusted data; do not visit or execute)
```json
{
  "domains": [],
  "ips": [],
  "hashes": [],
  "urls": []
}
```
