ai_security▌

2026-11-08

15 item(s) verified and swept on 2026-11-08, each citing a primary source.

NOTABLETOOL

OpenAI introduces GPT-6 Cyber, a vulnerability-finding model scoped to researcher-set boundaries

OpenAI launched GPT-6 Cyber, trained to find and validate exploitable vulnerabilities while staying within boundaries researchers set. OpenAI reports 48.5% on ExploitGym (up from 32.1% for GPT-5.6 Cyber) and says the model has already uncovered and validated exploits in security-critical codebases including Linux and Firefox.

NOTABLERED-TEAM

OpenAI partners with Hacktron to find vulnerabilities in its own systems

OpenAI announced a joint vulnerability research partnership with Hacktron focused on how OpenAI systems interact with third-party services, expanding Hacktron's access to OpenAI frontier models through Daybreak Red.

NOTABLEINCIDENT

Microsoft: threat actors use agentic AI tooling to find and abuse exposed secrets

Microsoft Security reports observing threat actors using agentic AI tooling to find exposed secrets and abuse them; the write-up was published on the Microsoft Security Blog on November 5, 2026.

NOTABLETOOL

vLLM: response amplification in the OpenAI-compatible validation error handler

GHSA-mwgq-9582-w2vf: vLLM's shared RequestValidationError handler rendered the full offending input for every validation error, so one malformed request generating many errors could serialize the same input repeatedly and amplify the HTTP 400 response. Fixed in 0.30.0; rated moderate.

NOTABLECVEPROMPT-INJECTION

CVE-2026-123653: DeepSeek-Reasonix fed untrusted document content straight into LLM prompting

GHSA-46pp-w572-hw78: discoverDocs() in esengine DeepSeek-Reasonix up to 2.52.0 passed input used for LLM prompting without neutralization, reachable remotely; the exploit is publicly disclosed and fixed in 2.53.0 (GHSA-rated low).

NOTABLECVE

CVE-2026-106020: Johnson Controls EasyIO Neo LLM prompting input handling allows search-order hijacking

GHSA-g936-836g-r2pg (from the NVD record) describes improper neutralization of input used for LLM prompting in Johnson Controls EasyIO Neo IO Manager and Tools Installer 1.1.0.0 allowing Search Order Hijacking; GitHub rates it high. Advisory text reads CNA-template-like — flagged for human review in the sweep log.

NOTABLEPROMPT-INJECTION

arXiv: benchmarking indirect prompt injection with task-tampering attacks

arXiv 2611.07640 benchmarks indirect prompt injection where injected content tampers with the agent's task rather than only overriding instructions.

NOTABLERED-TEAM

arXiv: Multimodal-SEA — AI agent security under multimodal social-engineering attacks

arXiv 2611.08281 benchmarks AI agent security under multimodal social-engineering attacks.

NOTABLERED-TEAM

arXiv: breaking TraintoDefend via adaptive neuron clustering in cryptanalytic model extraction

arXiv 2611.07370 breaks the TraintoDefend defense against model extraction using adaptive neuron clustering.

NOTABLERED-TEAM

arXiv: JailBound — FOL-guided jailbreak evaluation of LLM behavioral boundaries

arXiv 2611.07530 introduces JailBound, a first-order-logic-guided jailbreak evaluation method for exploring LLM behavioral boundaries.

NOTABLERED-TEAM

arXiv: 'Blind Compliance' — coercive control patterns in AI jailbreaking

arXiv 2611.08226 studies coercive-control dynamics behind 'Blind Compliance' in AI jailbreaking.

NOTABLERED-TEAM

arXiv: stress-testing chain-of-thought monitoring against covert misalignment

arXiv 2611.07586 stress-tests how well chain-of-thought monitoring detects covert misalignment.

NOTABLECVESUPPLY-CHAIN

CVE-2026-122160: AIlice agent framework loads models from untrusted control spheres

GHSA-6wf7-9xfh-x657: AIlice up to 0.4.0 passes base_model_name_or_path through its HuggingFace model wrapper (LoadModel_PEFT), including functionality from an untrusted control sphere; remotely attackable with a public PoC attached to the source report. GHSA rates it low.

NOTABLECVE

Cocos AI confidential-computing attestation flaws (CVE-2026-119344, CVE-2026-119345)

Cocos AI, a confidential-computing system for AI, accepted TEE attestation evidence without binding it to an allowed platform owner or machine identity (CVE-2026-119344) and checked attestation only once at TLS setup, trusting stale evidence after post-attestation tampering (CVE-2026-119345). Fixed in v0.9.0; reported in an ACM publication.

NOTABLECVE

CVE-2026-77604: code-graph-rag MCP read_file tool reads files outside the project root

GHSA-vvr2-h2jp-838m: the read_file MCP tool in code-graph-rag joined absolute file paths to project_root without containment validation on the paginated branch, letting an attacker read any UTF-8-readable file on the machine. Rated high; fixed in 0.0.345.

← 2026-11-09 · all days