Use case sources
Source reports and analysis supporting this case are listed below. AITBM interpretation and any indicative scoring are derived from this public evidence.
Sysdig recorded 10 attacker requests in eight minutes against its LMDeploy VLM-serving honeypot. Requests supplied image_url values through chat completions to probe IMDS and loopback services; unrestricted image loading reached addresses unavailable directly to the external client.
ASSESSED SYSTEM
Sysdig’s LMDeploy honeypot serving the stated InternLM-XComposer2 and InternVL2 vision-language models through its chat-completions image_url loader.
OUT OF SCOPE
Credential theft, successful destructive admin action, production third-party victims, model-assisted payload generation, and disputed fix metadata are outside the observed honeypot path.
Architecture: Standalone LLM / Generative AI (decision tree Q5) — The affected service is a named vision-language model inference endpoint; the report does not establish an agentic tool loop. Tier 2: Tier 2 representative public-facing service scenario; the original source does not establish every operator tier.
Documented attack or failure path
- An attacker sent image_url values to the public chat-completions endpoint of a VLM-serving LMDeploy honeypot.
- The image loader resolved attacker-supplied URLs against IMDS and loopback addresses.
- Sysdig observed probes and responses from its own honeypot, without evidence of credential theft or third-party victim compromise.
Layer 1 — Intrinsic Vulnerability Profile
Each sub-metric is placed on its five-level rubric by the evidence quoted beside it. Missing applicable evidence remains unknown. The displayed midpoint and interval are scenario values, not inferred control performance.
| Sub-metric | Score | Rubric basis | Evidence |
|---|---|---|---|
| Robustness (Ro) — scenario interval 0.00–1.00 (midpoint 0.50), Tier 2 MVT 0.50 indeterminate | |||
| Fairness (Fa) — scenario interval 0.00–1.00 (midpoint 0.50), Tier 2 MVT 0.55 indeterminate | |||
| Transparency (Tr) — scenario interval 0.00–1.00 (midpoint 0.50), Tier 2 MVT 0.50 indeterminate | |||
| Privacy (Pr) — scenario interval 0.00–1.00 (midpoint 0.50), Tier 2 MVT 0.55 indeterminate | |||
| Containment (Cn) — scenario interval 0.00–0.88 (midpoint 0.44), Tier 2 MVT 0.55 indeterminate | |||
| Cn-1Scope Enforcement | 0.00w 0.12 | The image-loading tool reached internal network locations outside the external user’s authorized scope, matching Cn-1 0.00. | Sysdig captured attacker image_url probes resolving to IMDS and loopback services from the LMDeploy VLM-serving honeypot.source: primary report |
Unknown, not N/A: 25 applicable sub-metrics lack admissible public evidence. The lower, midpoint, and upper scenarios evaluate each at 0.00, 0.50, and 1.00 respectively; no weight is redistributed merely because evidence is missing.
Layer 2 — Operational Risk Posture
| Dimension | Score | Justification |
|---|---|---|
| AaAutonomy Amplification | 0.25w 0.25 | Representative activity level for the documented bounded AI path; deployment-specific autonomy was not measured. |
| AsAttack Surface Exposure | 1.00w 0.35 | Representative exposure of the documented entry path; this is a scenario assumption and not a census of all operators. elevated |
| CpCascade Potential | 1.00w 0.20 | No verified System Dependency Graph or DGC is published, so framework worst-case Cp default applies.GDCP: worst-case default — no verified dependency graph published elevated |
| RfRemediation Feasibility | 0.50w 0.20 | A specific patch or configuration change is documented; repair feasibility is an inferred representative scenario, not measured recovery time. |
Nelevated = 2 (dimensions strictly above 0.75) → CRM = 1.15.
Compound Risk Alert. Two or more dimensions are simultaneously elevated (spec 3.2.2); architectural decomposition is recommended before deployment.
Layer 3 — Public-evidence confidence diagnostic
| Component | Score | Basis |
|---|---|---|
| Pc — Public provenance evidence | 0.25 | Minimal public provenance identifies the affected path but supplies no operator assessment pack, signed configuration inventory, or repeated metric battery. |
| Ec — Public evaluation coverage | 0.03 | coverage 0.04 (1 of 26 applicable sub-metrics) × independence 0.80 × fidelity 0.95. No Full, Standard, or Lite pathway is claimed for a retrospective article. |
| Tf — Public-evidence freshness | 0.25 | Evidence dated 2026-04-22; age 179 days on the workpaper reference date. Components: T_calendar 0.25 · C_monitor 0.60. Binding term: T_calendar. Evidence reference is 2026-10-18, 179 days after earliest substantive public evidence (2026-04-22). C_monitor uses the conservative Tier 2 public-evidence default because usable monitoring-coverage evidence is unavailable; it does not assert no deployment telemetry. No C_event cap is asserted from research or an advisory without a dated victim incident. No generic C_evidence cap is assigned from missing public inputs. Indicative ERS uses normalized ACI = 1.00 for comparison only. |
Public-evidence ACI = (Pc × Ec × Tf)1/3 = 0.12 — diagnostic status: Invalid as assessment-of-record evidence. It describes the evidence available to this case study, not the assurance of the underlying system, and is not inserted into the normalized-assurance scenario ERS.
Indicative ERS — normalized-assurance scenario
Published CVSS beside indicative ERS
Each published CVSS record is shown exactly as its scorer published it, one per vulnerability record, with the version, metric group and vector when stated, plus its source and access date. Records are never combined or inferred from a CVE's existence.
INDICATIVE ERS · NORMALIZED-ASSURANCE SCENARIO
4.8 midpoint, interval 1.3–8.2; Low-Moderate scenario band; ACI fixed at 1.00. The published CVSS records do not enter this calculation.
PUBLISHED CVSS · 1 record
CVE-2026-33626 CVSS 3.1 Base 7.5 High
| Published score | 7.5 |
|---|---|
| CVSS version | 3.1 |
| Metric group | Base |
| Severity | High |
| Vector | CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:N/A:N |
| Scored by | GitHub maintainer advisory |
| Source record | https://github.com/advisories/GHSA-6w67-hwm5-92mq (accessed 2026-10-18) |
| Affected component | LMDeploy vision image_url loader through 0.12.2 |
| Scope note | GitHub maintainer GHSA reports 7.5 CVSS 3.1 for the image URL SSRF path. |
| Source note | The advisory lists Patched versions None while its narrative says prior to 0.12.3; no current fix state is asserted. |
CVSS describes an individual vulnerability as its scorer published it. Indicative ERS describes the assessed AI deployment under the normalized-assurance scenario, so the two values are not direct equivalents and neither replaces the other.
AIDEFEND defences → AITBM sub-metrics
Identifiers are quoted as they appear on the AIDEFEND in Action brief or in the AIDEFEND SecureFlow defense map (audited AIDEFEND dataVersion 2026.10.15); the sub-metric mapping is AITBM's own, from the specification's AIDEFEND tables reconciled at catalogue data version 2026.10.15. AIDEFEND renumbers identifiers between releases, so the data version travels with every mapping and neither side's IDs should be cited without one. A mapping identifies a possible evidence route; a recommendation does not prove that the control was implemented or effective and receives no scoring credit by itself.
| Technique | Defence | Priority | Evidences |
|---|---|---|---|
| AID-E-001.001 | Root & Long-Lived Credential Object EvictionExact upstream page defense ID, verified against AIDEFEND dataVersion 2026.10.15. Presence supplies no score credit. | Candidate evidence route | Cn-5 |
| AID-H-003.010 | Deployed AI Software Vulnerability Remediation LifecycleExact upstream page defense ID, verified against AIDEFEND dataVersion 2026.10.15. Presence supplies no score credit. | Candidate evidence route | Ro-4 Ro-5 Tr-4 |
| AID-H-004.002 | Service & API AuthenticationExact upstream page defense ID, verified against AIDEFEND dataVersion 2026.10.15. Presence supplies no score credit. | Candidate evidence route | Tr-3 Cn-5 |
| AID-H-019.001 | URL Normalization & Allowlist FilteringExact upstream page defense ID, verified against AIDEFEND dataVersion 2026.10.15. Presence supplies no score credit. | Candidate evidence route | Cn-1 Cn-3 Cn-8 |
| AID-H-038.001 | Inference Runtime Listener & Protocol ConformanceExact upstream page defense ID, verified against AIDEFEND dataVersion 2026.10.15. Presence supplies no score credit. | Candidate evidence route | Ro-4 Cn-1 Cn-2 |
| AID-H-038.002 | Inference Runtime Privileged Operation AuthorizationExact upstream page defense ID, verified against AIDEFEND dataVersion 2026.10.15. Presence supplies no score credit. | Candidate evidence route | Ro-4 Cn-1 Cn-2 |
| AID-I-002.001 | Internal AI Network SegmentationExact upstream page defense ID, verified against AIDEFEND dataVersion 2026.10.15. Presence supplies no score credit. | Candidate evidence route | Cn-4 |
WHAT THIS CASE TEACHES
Sysdig recorded 10 attacker requests in eight minutes against its LMDeploy VLM-serving honeypot. Requests supplied image_url values through chat completions to probe IMDS and loopback services; unrestricted image loading reached addresses unavailable directly to the external client.
Sources: Official AIDEFEND in Action brief · Sysdig original LMDeploy honeypot observation, 2026-04-22 · LMDeploy maintainer GHSA and CVSS, 2026-04-20