Beyond IOCs: AI-enabled threat intelligence

Welcome to this week's Threat Source newsletter. The issue of AI in cybersecurity is often portrayed as a binary choice: either a force multiplier for our adversaries, or a tool bringing professional obsolescence. The reality is more nuanced. While AI certainly brings some advantage to attackers,...

pentest-agent-vs-llm-benchmark-effectiveness

Backbone or Backbone-Architecture? A controlled study of LLM age...

Interesting Paper Exploring Prompt Injection

This is a fascinating explotation of how LLMs fall for prompt injection attacks. It turns out that they learn to recognize the style of text in different role/instruction blocks, and not just the tags. Their conclusion: Role tags were a formatting trick that became the security architecture and t...

PT-2026-52617

Name of the Vulnerable Software and Affected Versions picklescan versions prior to 0.0.30 Description The software fails to detect malicious pickle files that invoke idlelib.pyshell.ModifiedInterpreter.runcode within reduce methods. This allows attackers to embed undetected code in pickle files...

Application of LLMs to Threat Assessment of Foreign Peacekeeping Missions

We present a novel approach for applying Large Language Models LLMs to threat assessment in the context of foreign peacekeeping missions. Building on the PINPOINT project and its use case, the EU Monitoring Mission in Georgia, we combine an interdisciplinary risk-model with OSINT-based media...

Dawn of the Apex Agentic Adversary

We are standing at the end of an era we never thought to mourn: the era of human-speed threats. For years, cybersecurity moved to a rhythm organizations could follow. A researcher found a bug, a CVE was cataloged, a vendor navigated a patch cycle, and weeks or even months later, a fix was deploye...

CrypFormBench: Benchmarking Formal Analysis Capability of Large Language Models for Cryptographic Schemes

Manual formal analysis of cryptographic schemes is labor-intensive and requires substantial expertise. While model-checking tools e.g., Scyther and Tamarin and computational-security tools e.g., CryptoVerif and EasyCrypt improve the automation of security proofs, they still rely on experts to...

CVE-2026-54013

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.6, Open WebUI patched SVG XSS in user profile images and webhook profile images but forgot to apply the same fix to model profile images. The ModelMeta class has no...

CVE-2026-54013

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.6, Open WebUI patched SVG XSS in user profile images and webhook profile images but forgot to apply the same fix to model profile images. The ModelMeta class has no...

CVE-2026-54013 Open WebUI: Stored XSS to Account Takeover via Model Profile Images in Open WebUI

Open WebUI is a self-hosted artificial intelligence platform designed to operate entirely offline. Prior to 0.9.6, Open WebUI patched SVG XSS in user profile images and webhook profile images but forgot to apply the same fix to model profile images. The ModelMeta class has no...

Poisoned Playbooks: Demystifying Knowledge Poisoning Effects on AI Security Agents

AI security agents increasingly rely on Retrieval-Augmented Generation RAG to use external security knowledge for vulnerability analysis and exploit reasoning. This creates a new risk: poisoned write-ups can be operationalized into incorrect exploit behavior. Yet, prior work on RAG poisoning has...

HelpBench: Assessing the Ability of LLMs to Provide Privacy, Safety, and Security Advice

This paper introduces HelpBench, a benchmark for assessing whether LLMs are capable of providing accurate help in response to questions about digital privacy, safety, and security. We curated 450 questions representing authentic user situations and developed rubrics for each question to evaluate...

Inside Crypter-As-A-Service: An Ecosystem Analysis of the Exploit.In Underground Forum Research Talks

Crypter-as-a-Service CraaS has become a key enabling layer of the contemporary malware economy by providing on-demand evasion capabilities through underground service markets. In this paper, we present a longitudinal characterization of the CraaS ecosystem on exploit.in, a major Russian-language...

PYSEC-2026-2300

vLLM is an inference and serving engine for large language models LLMs. Prior to 0.22.0, an assert-based security check in vLLM's activation function loading allows any unauthenticated attacker to achieve arbitrary code execution on the server by publishing a malicious HuggingFace model, when vLL...

Cleartext Storage of Sensitive Information

Overview Affected versions of this package are vulnerable to Cleartext Storage of Sensitive Information due to storing sensitive enrollment tokens in plaintext within the enrollmenttokens.token column of the database. An attacker can gain unauthorized access to host identities by reading the...

CVE-2026-53923

Summary of CVE-2026-53923 : The vulnerability affects vLLM (GGUF dequantize kernels) where integer truncation of tensor dimensions causes partially filled output tensors. From 0.5.5 up to 0.23.1rc0, the code allocates the full output tensor (torch::empty) but the CUDA kernel processes only a trun...

EUVD-2026-38400

vLLM is an inference and serving engine for large language models LLMs. From 0.5.5 until 0.23.1rc0, integer truncation of tensor dimensions in vLLM's GGUF dequantize kernels csrc/quantization/gguf/ggufkernel.cu causes partial tensor processing. The output tensor is allocated at full size via...

CVE-2026-56424 Broken access control in MISP core allows cross-organization unauthorized modification or deletion of analyst data, event reports, collections, templates, and decaying models

MISP core contained multiple broken access-control flaws where authorization checks were performed against the wrong entity, or where ownership/editability checks were missing on write paths. In affected subsystems, a lower-privileged authenticated user with the relevant feature permission could...

CVE-2026-56424

CVE-2026-56424 affects MISP core and describes multiple broken access-control flaws where authorization checks target the wrong entity or where ownership checks are missing on write paths. In affected subsystems, a lower-privileged authenticated user with relevant feature permissions could cause ...

CVE-2026-56424 Broken access control in MISP core allows cross-organization unauthorized modification or deletion of analyst data, event reports, collections, templates, and decaying models

MISP core contained multiple broken access-control flaws where authorization checks were performed against the wrong entity, or where ownership/editability checks were missing on write paths. In affected subsystems, a lower-privileged authenticated user with the relevant feature permission could...