AI 安全
30 articles about "AI 安全".
What Is PHP2Shell? The PHP Flaws Under 70% of the Web, and the Three Things Your Site Needs Now
What is PHP2Shell? Two PHP flaws (CVE-2026-109618, 109619, CVSS 9.2) that may allow unauthenticated code execution, fixed in 8.5.12, 8.4.27, 8.3.36, 8.2.35.
What Is Prompt Injection? How It Reaches an AI Assistant Through Documents and Tools, and Three Layers of Defense (2026)
What is prompt injection? How hidden instructions reach an AI assistant through documents, web pages and tools, and three layers of defense.
ChatGPT Privacy Center Is a Map, Not a New Switch: Reading It Through Four Questions
What is ChatGPT Privacy Center? A map, not a new switch: it changes no settings and deletes no data. Read through default, scope, retention, revocation.
Regex Can't Tell "Said It" From "Warned Against It": False Positives, Misses, and Our Overturned Fix in LLM Output Scanning
Matching LLM replies with regex to catch XSS or SQL injection strings also fails replies that name the hazard in order to warn against it: on an unmerged open-source Giskard PR, all 5 such advice sentences tripped the check. We proposed anchoring the patterns to line start and false positives fell to 0. Then the PR author added one ordinary phrasing, a lead-in on the same line as the payload, and the anchor missed 12 of 24 compliant replies. This post covers why the anchor only held on our sample, how to choose between over-reporting and under-reporting, and why our own output scanner has the same limit.
AI Skills Are an Injection Path Too: Scanning 954 Popular Skills, and an Eighth Question Before Buying an AI System
AI skills are an injection path too: a 2026 paper scanned 954 agent skills, confirmed 84 of 100 sampled findings, and landed 13 of 30 attacks.
ChatGPT's Terms Treat Your Face as "Content": Reading OpenAI's Terms on Uploaded Photos
A paragraph-by-paragraph reading of OpenAI's official terms. The Terms of Use file uploaded images under "Input", the same category as typed text; ownership stays in your name while usage rights also go to OpenAI; in the Service Terms passage on likeness, all six verbs have "you" as their subject; and the training help page explicitly separates services for individuals from business products. Nowhere in the terms does the face get a section of its own. Every quotation was checked against the official pages as they read at the time of writing.
Uploading a Selfie to ChatGPT Is Not a Filter: Your Face Is a Credential You Cannot Reset
Uploading a selfie to ChatGPT is not a filter: a face is a credential you cannot reset, and opting out later does not pull it from training.
If Your Product Takes Photo Uploads, You Are Already Building a Biometric System: Four Questions for Every Image Endpoint
For people building AI products: once there is a path where a user uploads a photo and another face is generated from it, the data category is already identity material, whether the feature copy calls it a dress-up or a filter. Every upload endpoint that takes images needs four answers fixed before launch: default, scope, retention and revocation. Includes a full checklist, plus the two paths in our own products that touch faces and the work we have not finished yet.
What Is GPT-6 Astra? Capabilities, Pricing, API Migration, and Five Safety Numbers to Read Before You Hand It an Agent
What is GPT-6 Astra? OpenAI's September 2026 model: pricing, API migration from GPT-5.6, and five system card numbers that matter before you give it an agent.
The Scanner Had the Bug It Was Looking For: Auditing Someone Else's Invisible-Character List, Then Our Own
Anthropic's commerce-agents blueprint strips invisible characters with a hand-written list. A Unicode category sweep found 34 Cf code points and four invisible non-Cf characters missing from it, enough to keep a fence label or a role word intact through sanitizing. Then we pointed the same probe at our own code and all three of our scanners missed the same set. On why enumerations expire, how to grade a finding honestly, and how binding rules to wording instead of concepts makes you systematically underrate the systems that got it right.
Why Agentic AI Attack Testing Shouldn't Be One Class Per Attack: The Vector / Framing / Scorer Decomposition
In an agent pipeline the same malicious payload can enter as a tool result, a retrieved document, or a sub-agent message. Write one attack class per entry point and your test harness explodes. This is the three-axis model we posted publicly in microsoft/PyRIT: injection vector is data, framing is a transform, the scorer is the verdict, and why an attack is a placement, not a message.
The July 2026 MCP Server Auth Epidemic: 12 Projects, 19 Advisories, and the Reference SDK Itself
In three weeks of July 2026, at least 12 MCP projects, including the official MCP Python SDK itself, shipped 19 security advisories, all pointing at the same thing: authentication and origin validation treated as optional. This isn't a run of bad luck, it's a structural gap. A source-verified inventory, the five recurring classes and their root cause, and the checklist to run before you connect any MCP server.
OWASP Agentic Top 10 (ASI-01 to ASI-10) Explained for 2026: Real Scenarios, Detection and a Fix Checklist for Each Risk
OWASP Agentic Top 10 (ASI-01 to ASI-10) risk by risk: what it is, a real scenario, how to scan for it with open-source tools, and a fix checklist to copy.
OWASP LLM Top 10 Explained (Official 2025 Edition, 2026 Status): How It Differs from the Agentic ASI Top 10 and How Developers Should Defend
OWASP LLM Top 10 2026: the latest official list is still the 2025 edition; what is new is the Agentic ASI attack surface. All 10 risks and three defense layers.
AI Agent Output Guardrails: How to Stop API Key Leaks Before Your Agent Posts
We sell AI guardrails, yet our own agent was posting publicly with no output checks and could have leaked API keys. How to build an egress guardrail, and how to decide what to hard-block and what only gets a soft warning.
From 6 to 21: The Crypto AI Agent Incident Tracker Goes Live ($52M of Documented Loss)
The 6 incidents from an earlier analysis, expanded to 21 today. $52M total documented loss. Structured data, open-source repo, public page. Built in-flight.
Six Crypto AI Agent Heists: What Static Prompt Analysis Catches, What It Doesn't
An honest root-cause analysis of six prompt-injection incidents that drained crypto AI agents, and a measured assessment of what prompt-defense-audit can and cannot catch.
We Audited 7 Official MCP Servers: 6 Got F
Ran prompt-defense-audit against the 7 official servers in modelcontextprotocol/servers: 12-vector check, OWASP LLM Top 10 mapping. Result: 6 servers scored F, 8 defense vectors at 100% gap rate. Cross-referenced from modelcontextprotocol/servers#3537.
Cisco Merged My PR in 51 Minutes: Why Prompt Defense Is the Next SQL Injection
AI agents and chatbots are growing exponentially, foundation models update every three months, but 78% of the system prompts leaked online have zero defense lines. From one casual scan to Cisco merging in 51 minutes and Microsoft assigning me an issue: the four months between.
OWASP Agentic Top 10: What Every AI Developer Needs to Know in 2026
OWASP released its Top 10 security risks for AI agent applications in 2026. Each risk broken down with data from scanning 1,646 leaked public system prompts.
One Line to Block 92% of Prompt Injection Attacks
Our Discord AI assistant gets attacked every few days. After scanning 1,646 system prompts leaked online, we built a one-liner defense tool.
We Built Lighthouse for AI Agents: One Command, 25-Vector Security Audit
66% of MCP servers have security findings, but nobody runs a security scan before deploying AI agents. We built ultraprobe: zero deps, zero cost, under 1 second. Its prompt defense scanning technique was merged into Cisco AI Defense's MCP Scanner (PR #146).
12 Submissions, 0 Merges: What I Learned Contributing to Open Source AI Security
We submitted contributions to Cisco, Microsoft, OWASP, and 9 other open source projects. All rejected or ignored. Here's how we went from 0/12 to our first merge.
We Defined an AI Security Standard: AASS v1.0, We Don't Sell Security, We Define It
AI Application Security Standard (AASS) is an open standard published by Ultra Lab that covers AI system defense, website AI visibility, and data protection in a single framework. All tools free and open source.
We Scanned 1,646 Leaked AI System Prompts. Here's What We Found.
We ran our prompt defense scanner against 1,646 system prompts leaked online from ChatGPT, Claude, Grok, Cursor, Perplexity, and 1,300+ custom GPTs. 97.8% have no indirect injection defense. Average score: 36/100.
Prompt Injection Isn't Your Biggest Risk: 517 AI System Prompts Scanned, 11 Undefended Attack Vectors
Prompt Injection is 1 of 12 LLM attack vectors. Ultra Lab's UltraProbe scan of 517 prompts (sample unpublished) found 3.2 of 12 defended. Examples and fixes.
78.3% Score F: Prompt Defense Gap Data from 1,646 Leaked AI System Prompts
We scanned 1,646 system prompts leaked from GPT Store, ChatGPT, Claude, Cursor and others. The average score was 36/100 and 78.3% scored F. This post uses the data to show how serious each OWASP Agentic Top 10 risk is in the real world.
We Open-Sourced Our Prompt Defense Scanner: 200 Lines of Regex That Replace an LLM
Most AI security tools use LLMs to check LLMs. We built a deterministic prompt defense scanner: pure regex, under 1ms, zero cost (12 defense vectors at launch, 17 today). Here's why regex beats AI for this job, and how you can use it today.
How We Defend AI Against Comment Attacks: 5-Layer Prompt Defense in Production
When your AI auto-replies to hundreds of comments daily, Prompt Injection isn't theoretical: it's happening every day. This is the 5-layer defense architecture we validated across 27 accounts.
UltraProbe, a Free AI Security Scanner, Is Live: Find Your LLM App's Vulnerabilities in 5 Seconds
90% of AI systems are vulnerable to Prompt Injection, yet most developers have no idea. Ultra Lab launches the completely free UltraProbe, mapped to four OWASP LLM Top 10 (2025) categories: LLM01, LLM02, LLM05 and LLM10, making AI security testing accessible to everyone, not just enterprises.