Topics
Browse posts by category and tag — every topic we cover, with the latest pieces under each.
Tags
- #prompt-injection 18
- #llm-security 16
- #supply-chain 12
- #cve 10
- #ai-security 7
- #nist 7
- #owasp 7
- #adversarial-ml 6
- #data-poisoning 6
- #cisa 5
- #enterprise-ai 5
- #openai 5
- #indirect-injection 4
- #red-teaming 4
- #agent-security 3
- #bec 3
- #chatgpt 3
- #deepfakes 3
- #detection 3
- #generative-ai 3
- #huggingface 3
- #jailbreaking 3
- #mlsecops 3
- #pickle 3
- #primer 3
- #rag 3
- #safetensors 3
- #social-engineering 3
- #threat-intelligence 3
- #2026 2
- #advisory 2
- #ai-security-alert 2
- #analysis 2
- #api-security 2
- #attribution 2
- #breach 2
- #compliance 2
- #cve-roundup 2
- #data-exfiltration 2
- #deserialization 2
- #dlp 2
- #enterprise-security 2
- #federal 2
- #governance 2
- #hugging-face 2
- #incident-response 2
- #incidents 2
- #knowledge-base 2
- #machine-learning-security 2
- #may-2026 2
- #methodology 2
- #model-extraction 2
- #model-hub 2
- #model-theft 2
- #monitoring 2
- #patch-management 2
- #poisoning 2
- #ransomware 2
- #red-team 2
- #synthetic-media 2
- #system-prompt 2
- #threat-brief 2
- #vector-database 2
- #vllm 2
- #vulnerability-disclosure 2
- #access-control 1
- #active-exploitation 1
- #adversarial-prompting 1
- #adversarial-testing 1
- #agentic 1
- #agentic-ai 1
- #ai-assisted-attack 1
- #ai-governance 1
- #ai-policy 1
- #ai-risk-management 1
- #ai-risks 1
- #ai-rmf 1
- #ai-sbom 1
- #ai-security-guidance 1
- #ai-security-tools 1
- #ai-security-update 1
- #ai-vulnerabilities 1
- #alignment 1
- #alignment-bypass 1
- #anthropic 1
- #api-keys 1
- #assessment 1
- #audit 1
- #autonomous-ai 1
- #aws 1
- #backdoor 1
- #benchmarking 1
- #bing 1
- #bug-bounty 1
- #checklist 1
- #cisa-kev 1
- #claude 1
- #cms 1
- #code-execution 1
- #confidentiality 1
- #containment 1
- #copilot 1
- #credential-leak 1
- #cryptography 1
- #data-leak 1
- #data-privacy 1
- #defense 1
- #digest 1
- #disclosure 1
- #dprk-it-workers 1
- #email 1
- #eu-ai-act 1
- #evaluation 1
- #evasion 1
- #forensics 1
- #framework 1
- #gandcrab 1
- #garak 1
- #github 1
- #govcloud 1
- #gpt-4 1
- #gradio 1
- #identity-fraud 1
- #image-collision 1
- #incident 1
- #incident-analysis 1
- #incident-catalog 1
- #inference-attacks 1
- #infrastructure 1
- #infrastructure-security 1
- #input-filtering 1
- #intellectual-property 1
- #jailbreak 1
- #kev 1
- #known-exploited 1
- #langchain 1
- #langchain-chatchat 1
- #langflow 1
- #law-enforcement 1
- #linux-kernel 1
- #litellm 1
- #llm-abuse 1
- #llm-agents 1
- #llm-guard 1
- #llm-scanner 1
- #llm-security-alert 1
- #llm-serving 1
- #local-privilege-escalation 1
- #logging 1
- #machine-learning 1
- #malicious-models 1
- #md5 1
- #membership-inference 1
- #metrics 1
- #microsoft-365 1
- #mitre-atlas 1
- #ml-infrastructure 1
- #ml-pipeline 1
- #mlops 1
- #mlsec 1
- #model-format 1
- #model-poisoning 1
- #model-security 1
- #model-serving 1
- #nvidia 1
- #ollama 1
- #onnx 1
- #open-source 1
- #outlook 1
- #output-filtering 1
- #pentesting 1
- #phishing 1
- #php 1
- #pii 1
- #playbook 1
- #privacy 1
- #pytorch 1
- #q1-2026 1
- #raas 1
- #remote-code-execution 1
- #retrieval 1
- #revil 1
- #risk-framework 1
- #risk-management 1
- #safety 1
- #samsung 1
- #secure-by-design 1
- #sglang 1
- #shadow-ai 1
- #siem 1
- #slack 1
- #spearphishing 1
- #threat-alert 1
- #threat-intel 1
- #tool-review 1
- #tool-use 1
- #tools 1
- #top-10 1
- #training-data 1
- #triton 1
- #vishing 1
- #vision-llm 1
- #vulnerability 1
- #vulnerability-scanner 1
- #weak-hash 1
- #weekly 1
- #wild 1
Categories
analysis 13 posts
- The NIST AI Risk Management Framework: A Practitioner's Field GuideWhat the NIST AI RMF actually asks you to do — the GOVERN, MAP, MEASURE, MANAGE functions, the Generative AI Profile's 12 risk categories, and how a
- CISA AI Security Guidance: What Organizations Need in 2026A breakdown of CISA's published AI security guidance — what it covers, what it requires, and how organizations should operationalize it.
- Generative AI Risks: A Practical Taxonomy for Security TeamsGenerative AI risks span prompt injection, data poisoning, supply chain vulnerabilities, hallucination, and governance failures.
- How to Track AI Security Alerts: CISA, NIST, and Vendor FeedsA practical guide to the official and community sources for AI security alerts — what each publishes, how frequently, and how to integrate them into a
- LLM Security Alerts: Monitoring, Detection, and ResponseA practical guide to setting up LLM security alerting — what to monitor, what alert patterns indicate compromise or attack, how to triage LLM security
- LLM Security Risks: A Practitioner's Field Guide for 2025A comprehensive breakdown of LLM security risks — prompt injection, supply chain poisoning, excessive agency, and model extraction — with mitigation
disclosure 6 posts
- LLM Security Risks: The 2025 Threat Landscape for AI DeploymentsA practitioner breakdown of LLM security risks covering the OWASP Top 10 for LLM Applications 2025, MITRE ATLAS AI attack tactics, and concrete mitigation
- Generative AI Risks: A Technical Reference for Security TeamsA practitioner-focused breakdown of generative AI risks mapped against NIST AI 600-1 and the OWASP Top 10 for LLMs — prompt injection, data poisoning
- Machine Learning Security: Attack Taxonomy, CVEs, and DefensesA technical overview of machine learning security threats in 2026: NIST's adversarial ML taxonomy, MITRE ATLAS attack classes, the CVE-2025-62164 vLLM
- OpenAI Security: Bug Bounties, CVE Disclosure, Mixpanel BreachA practitioner's overview of OpenAI security in 2026: their bug bounty program, CNA status, the November 2025 Mixpanel breach, and what security teams
- ChatGPT Security: Patched Flaws, Persistent Gaps, Unsolved RisksA technical review of ChatGPT security vulnerabilities disclosed in 2025-2026: DNS-based data exfiltration, ZombieAgent prompt injection bypass, Codex
- AI Security: Attack Categories, Defense Gaps, and How to RespondA practitioner guide to the four core attack categories against AI/ML systems — from adversarial inputs to supply chain compromise — with mitigation
deep-dive 5 posts
- The Gentlemen Ransomware: AI-Assisted Development, 332 Victims, and the OSINT Trail That Exposed a Russian AdminSecurity researchers have identified the suspected administrator of The Gentlemen, the world's second-most-active RaaS by victim count, as a Russian
- RAG Poisoning: How Retrieval-Augmented Systems Get CompromisedRAG systems inherit all the vulnerabilities of LLMs and add a new one: the retrieval corpus. Injecting malicious content into retrieved sources can hijack
- Model Extraction Attacks: How Adversaries Steal AI via the APIModel extraction attacks reconstruct proprietary AI models by querying their public APIs. Here's how they work, what has been demonstrated against real
- LLM Supply Chain Poisoning: Training Data Attacks and BackdoorsTraining data poisoning and model supply chain attacks are among the hardest AI threats to detect. This post explains how they work, what public research
- Germany Names UNKN: What the BKA's REvil and GandCrab Dox BuysGermany's BKA has put a name and a face to UNKN, the operator behind GandCrab and REvil. Russia will not extradite, but the wanted notice is doing other
cve 4 posts
- CVE-2026-7845: Hash Collision in Langchain-ChatchatA weak-hash flaw in Langchain-Chatchat up to 0.3.1.3 lets an adjacent attacker overwrite pasted images by colliding MD5 hashes computed from PIL.Image.
- MetInfo CMS CVE-2026-29014 Exploited in the Wild for RCEA critical unauthenticated PHP code injection flaw in MetInfo CMS 7.9–8.1 (CVSS 9.8) is under active exploitation. Patch to the April 7 release immediately.
- CISA Adds Exploited Linux Kernel LPE CVE-2026-31431 to KEVA local privilege escalation flaw in the Linux kernel's AEAD crypto interface has been added to CISA's KEV catalog after active exploitation.
- CVE-2026-7669: Deserialization Flaw in SGLang's Tokenizer LoaderA medium-severity deserialization bug in SGLang's get_tokenizer routine affects all releases up to 0.5.9. The vendor has not responded to the disclosure
guide 4 posts
- Building an AI Security Incident Response PlanA practical incident response plan for AI systems — what a prompt-injection or model-extraction incident looks like, how the NIST SP 800-61r3 / CSF 2.
- ChatGPT Security: Risks, Controls, and How to Use It SafelyA practitioner's guide to ChatGPT security in 2026: how OpenAI protects enterprise data, where prompt injection and account-takeover risks live, and the
- Machine Learning Security Across the Pipeline: Data to DeploymentMachine learning security vulnerabilities enter at every stage — data ingestion, model training, artifact storage, and inference.
- Machine Learning Security: Threats, Frameworks, and DefensesA practitioner's reference for machine learning security: the canonical attack categories, the frameworks that catalog them (NIST AI 100-2, OWASP ML Top
threat-brief 4 posts
- Deepfake Cybersecurity: Detection Methods and Practical DefensesFrom the FBI's May 2025 warning on AI voice attacks targeting US officials to NIST's synthetic content framework, here is what detection technology
- Deepfake Cybersecurity: Attack Vectors and Defenses for 2026Deepfake cybersecurity has moved from theoretical risk to documented billion-dollar loss category. Here's what the attack surface looks like, why
- AI-Generated Phishing and the Collapse of Spearphishing CostCrafting a convincing, personalized phishing email once required hours of research per target. Large language models have reduced that cost to seconds.
- Data Poisoning in RAG Systems: A 2026 Threat BriefingAttackers are actively poisoning retrieval-augmented generation knowledge bases in enterprise deployments. This briefing documents the current threat
incident 3 posts
- Deepfake Cybersecurity: Five Confirmed Cases and the PatternsA working catalog of confirmed deepfake cybersecurity incidents from 2024 and 2025 — from Arup's $25M loss to Ferrari's averted scam — and the controls
- The Samsung ChatGPT Data Leak: Lessons for Enterprise AIIn 2023, Samsung employees leaked proprietary source code and meeting notes through ChatGPT. This incident defined a category of enterprise AI risk that
- How System Prompt Leaks Happen: Techniques and DefensesPrompt injection attacks that expose system prompts are one of the most common real-world LLM exploits. This post covers the mechanics of system prompt
incident-review 3 posts
- Compromised Models on Hugging Face: Pickle Exploits in the HubMalicious actors have uploaded model files to Hugging Face containing pickle payloads that execute code on download.
- Hugging Face Security Incidents: Malicious Models and Token TheftA review of documented security incidents on the Hugging Face platform, including malicious model uploads, the 2024 Spaces infrastructure breach, and the
- Prompt Injection via Email: How AI Agents Get HijackedEmail is the highest-volume source of untrusted content in enterprise environments — and it's now being fed directly into AI agents.
primer 3 posts
- AI System Security Audit Checklist for 2026A practical audit checklist for AI systems covering model inputs, training pipeline, outputs, access control, logging, and red-team requirements.
- Jailbreaking vs Prompt Injection: Not the Same AttackSecurity practitioners conflate jailbreaking and prompt injection constantly. They are distinct attack classes with different threat actors, different
- What Is Adversarial ML? A Practitioner's PrimerA practitioner-focused introduction to adversarial machine learning: evasion, poisoning, and inference attacks, why they matter in production, key papers
advisory 2 posts
- Model File Format Flaws: Pickle, ONNX, and SafeTensorsUnsafe deserialization in PyTorch's pickle-based format has enabled malicious model distribution for years. This post explains how pickle exploitation
- Shadow AI: Ungoverned LLM API Keys and Data Exfiltration RiskEmployees using personal Claude, OpenAI, and Gemini API keys for work tasks bypass corporate DLP controls and send sensitive business data to external
cve-roundup 2 posts
- CVE Roundup: AI/ML Infrastructure Vulnerabilities — Q1 2026A quarterly review of critical CVEs disclosed in Q1 2026 affecting model serving infrastructure: vLLM, NVIDIA Triton Inference Server, Gradio, LangChain
- AI/ML CVE Roundup: May 2026 — What Got PatchedA summary of AI and ML-adjacent CVEs disclosed in early–mid 2026 across model serving frameworks, LLM API gateways, agent SDKs, and ML training libraries.
methodology 2 posts
- A Practical Guide to AI Red-Teaming for Security TeamsRed-teaming LLMs requires different skills and methodology than traditional network or application penetration testing.
- How to Benchmark AI Security Tools: A 2026 MethodologyChoosing an AI security tool without a structured evaluation methodology is expensive guesswork. This guide covers the metrics that matter, the pitfalls