AI Sec
Home Entries Topics
☰
Tools Attack Atlas Glossary Resources About this index
Offensive Sec network
Adversarial ML AI Attacks JailbreakDB Jailbreaks FYI ML CVEs

New here?

Start here.

aisec.blog is the flagship of a 26-site network covering AI security end-to-end — from the prompt-injection PoC on offense, to the guardrail that catches it on defense, to the incident desk that logs it the day it lands in production. This page is the front door: what we cover, what to read first, and how to keep up.

Why this exists

AI security is fragmenting into a dozen sub-disciplines — prompt injection, jailbreaks, model supply-chain CVEs, content moderation, eval reliability, ML observability, AI policy. No single publication can cover all of it without going shallow. The AI Sec network solves that with 26 specialist sites, each owned by one editorial lane, cross-linked when stories genuinely overlap.

aisec.blog is the offensive-security flagship. Sister sites cover defensive engineering, production LLMOps, incident tracking, AI policy, and curated news. Pick the lanes that match your job; ignore the rest.

Read these first

Five pillar posts per cluster — curated picks that introduce the lane without requiring prior context. If you only read one cluster, read the one that matches your role.

Offensive

7 sites

Prompt injection, jailbreaks, agent and tool-use exploitation, adversarial ML.

  1. LLM Security: A Practitioner's Map of the Attack Surface · aisec.blog
  2. Direct vs. Indirect Prompt Injection: Threat Models, Attack Surface, and Defense Differences · aisec.blog
  3. Why Your Prompt Injection Guardrails Fail: A Practitioner's Tour of Bypass Classes · aisec.blog
  4. LLM Jailbreak: Attack Taxonomy, Live Techniques, and Defense Reality · aisec.blog
  5. AI Red Teaming Hub: Your Guide to Offensive AI Security on aisec.blog · aisec.blog

Defensive

7 sites

Guardrails, content filters, detection engineering, and the tools that actually hold under fire.

  1. Guardrails that actually hold — what the stack looks like in 2026 · guardml.io
  2. AI Defense: hardening the LLM call path end-to-end · aidefense.dev
  3. AI Moderation Tools: side-by-side reviews of content-safety APIs · aimoderationtools.com
  4. Best LLM Scanners: comparing prompt-injection detection tools · bestllmscanners.com
  5. AI Sec Bench: benchmarks for AI security tooling · aisecbench.com

Engineering

5 sites

Production LLM ops — evals, observability, drift, cost, latency. ML monitoring done with numbers.

  1. SentryML — production model monitoring and drift · sentryml.com
  2. LLMOps Report: evals, latency, and cost in production · llmops.report
  3. ML Observe: drift, debugging, and the metrics that matter · mlobserve.com
  4. ML Monitoring Report: reliability for production ML · mlmonitoring.report
  5. MLOps Platforms: comparisons, not vendor decks · mlopsplatforms.com

News

3 sites

Daily and weekly news desks covering AI security across vendors, research, and disclosures.

  1. Tech Sentinel — daily cybersecurity news · techsentinel.news
  2. AI Sec Digest — curated AI security news, daily · aisecdigest.com
  3. AI Sec Weekly — Friday digest · aisecweekly.com
  4. AI Sec Digest: this week in prompt-injection research · aisecdigest.com
  5. Tech Sentinel: vulnerabilities making news · techsentinel.news

Incidents

2 sites

Dated, sourced trackers for AI/ML incidents, breaches, model failures, and disclosed CVEs.

  1. AI Alert — incident & vulnerability tracker · ai-alert.org
  2. AI Incidents — dated, sourced log of AI failures · aiincidents.org
  3. AI Incident Logging Methodology · aiincidents.org
  4. Live incidents tracker (top 5 latest) · aisec.blog
  5. ML CVEs — the AI/ML supply-chain CVE feed · mlcves.com

Policy

2 sites

EU AI Act, NIST AI RMF, US state law, GDPR × AI, and regulator action — read article-by-article.

  1. NeuralWatch — EU AI Act, NIST AI RMF, US state law tracking · neuralwatch.org
  2. AI Privacy Report — GDPR × AI enforcement · aiprivacy.report
  3. NeuralWatch: regulator actions worth reading · neuralwatch.org
  4. AI Privacy Report: compliance for AI products · aiprivacy.report
  5. NeuralWatch: litigation tracker · neuralwatch.org

The network

Twenty-six publications, grouped by lane. Each is operated independently with its own editorial voice and primary sources.

Cluster Site What it covers
Offensive AI Sec aisec.blog Offensive AI security — prompt injection, jailbreaks, agent exploitation, red team writeups.
· Jailbreaks FYI jailbreaks.fyi Working LLM jailbreak techniques, sourced and dated.
· JailbreakDB jailbreakdb.com An indexed catalog of working LLM jailbreak techniques.
· AI Attacks aiattacks.dev Practitioner-grade AI red team techniques and tooling.
· Prompt Injection Report promptinjection.report Prompt injection PoCs, taxonomy, and primary sources.
· Adversarial ML adversarialml.dev Working adversarial ML — exploits, defenses, and the gap between.
· ML CVEs mlcves.com CVEs in ML libraries, frameworks, and the AI/ML supply chain.
Defensive GuardML guardml.io Defensive AI — guardrails, content filters, model defenses, safe deployment.
· AI Defense aidefense.dev Defensive AI engineering — guardrails, hardening, response.
· AI Moderation Tools aimoderationtools.com Honest reviews and benchmarks of AI content-moderation tooling.
· AI Sec Reviews aisecreviews.com Reviews of AI security products and platforms.
· Best AI Security Tools bestaisecuritytools.com Comparing the AI security tooling landscape, with numbers.
· Best LLM Scanners bestllmscanners.com Comparing LLM security scanners and detection tools.
· AI Sec Bench aisecbench.com Benchmarks and evaluations of AI security tools.
Engineering SentryML sentryml.com ML observability & MLOps — model monitoring, drift detection, debugging in production.
· LLMOps Report llmops.report Operating LLMs in production — eval, observability, cost, latency.
· ML Observe mlobserve.com ML observability deep dives — drift, debugging, monitoring.
· ML Monitoring Report mlmonitoring.report Production ML monitoring, drift, and reliability.
· MLOps Platforms mlopsplatforms.com Honest reviews and comparisons of MLOps platforms.
News Tech Sentinel techsentinel.news Cybersecurity news, daily — breaches, CVEs, ransomware, threat actors, and the patches that matter.
· AI Sec Digest aisecdigest.com Curated AI security news, daily.
· AI Sec Weekly aisecweekly.com Weekly digest of AI security news and analysis.
Incidents AI Alert ai-alert.org AI incidents and vulnerabilities — tracked, sourced, dated.
· AI Incidents aiincidents.org AI incidents, model failures, and adversarial-use cases — dated and sourced.
Policy NeuralWatch neuralwatch.org AI policy and ethics watchdog — regulation, accountability, governance.
· AI Privacy Report aiprivacy.report AI privacy regulation, compliance, and enforcement, sourced.

Keep up

One newsletter covers the whole network. Pick the clusters that match your job — change anytime, unsubscribe in one click.

  • Weekly digest, Friday afternoon UTC — one email
  • Optional daily news roundup (Mon–Fri) for news-cluster subscribers
  • No tracking pixels, no surveillance, no data sale
Subscribe to the newsletter

For practitioners