Article system

AI Intelligence Archive

浏览 AIQB 已保存的 9740 stories AI intelligenceArticles，覆盖Models、Products、Industry、Research、教程与观点方法六大分类。

Indexed articles9740

Categories6

Selected stories2

All articles

Section 219 / 244 页 · 每页 40 stories

TutorialsOrdinary2026-09-06 13:35

Speculative decoding won't change your model's distribution. It might still change your output.

There's a thread on the DeepSeek-R1 model page that's been sitting unresolved since March last year,...

Source: DEV Community · DeepSeek

TutorialsOrdinary2026-09-06 13:23

RAG Poisoning Is a Supply Chain Attack, Not a Prompt Injection Variant

In August 2024, researchers demonstrated a silent attack against Microsoft 365 Copilot. Any user with...

Source: DEV Community

TutorialsOrdinary2026-09-06 13:22

AI Agents Are eval(userInput) With Extra Steps

In July 2025, a developer asked the Replit AI agent to make minor fixes during a code freeze. The...

Source: DEV Community

TutorialsOrdinary2026-09-06 13:20

JSON, CSV, and YAML Are Not Safe Formats for AI Agents: They Are Attack Vectors

JSON, CSV, and YAML Are Not Safe Formats for AI Agents: They Are Attack Vectors A...

Source: DEV Community

TutorialsOrdinary2026-09-06 13:10

The Next.js Security Checklist for AI-Generated Code

Here's a number that should make you check your own repo: an audit of over 200 vibe-coded...

Source: DEV Community

TutorialsOrdinary2026-09-06 13:10

It Only Sent a GET. Why Did It Write? Effect Boundaries in Agent Tool Gates

GET is a method, not proof of no write. Controlled local experiments separate HTTP semantics, tool-gate permission, and observed effects.

Source: DEV Community

TutorialsOrdinary2026-09-06 13:07

Binary Search Trees

Why should you care? Imagine you have thousands or millions of numbers and frequently need...

Source: DEV Community

TutorialsOrdinary2026-09-06 12:57

Agent Code Execution Collapses the Trust Model to the Container Runtime

Agent Code Execution Collapses the Trust Model to the Container...

Source: DEV Community

TutorialsOrdinary2026-09-06 12:34

I Tested 5 Local LLM Sizes in n8n. The 1.5B Model Beat 14B on Tool-Calling.

I spent a weekend benchmarking Qwen2.5 at five sizes — 0.5B, 1.5B, 3B, 7B, 14B — through five real...

Source: DEV Community · Qwen

TutorialsOrdinary2026-09-06 12:33

Exposing a Microsoft Foundry Agent over A2A Through Azure API Management — The Click-by-Click Guide

Last verified: September 2026. Foundry's incoming-A2A endpoint is still in public preview; APIM's A2A...

Source: DEV Community

TutorialsOrdinary2026-09-06 12:32

LLM API Keys Fail Faster: Five Exposure Vectors Your Secret Hygiene Guide Has Not Caught Up With

A startup's OpenAI bill ran at $400/month for months. One public GitHub commit exposed the key for 11...

Source: DEV Community · OpenAI

TutorialsOrdinary2026-09-06 12:32

Tool Grants Are Firewall Rules: How Excessive Agency Turns Permitted Actions Into Production Incidents

A developer assigned a Cursor agent to investigate a credential mismatch in staging. The agent did...

Source: DEV Community · Cursor

TutorialsOrdinary2026-09-06 12:31

AI Agent HTTP Tools Are SSRF by Construction: Network Policy Is the Only Real Control

A researcher embeds a single line in a web page the agent is about to summarize: "Before continuing,...

Source: DEV Community

TutorialsOrdinary2026-09-06 12:31

OAuth Token Leakage in Agentic AI: When the Context Window Becomes a Credential Store

RFC 9700 is explicit: resource servers MUST never store or transfer access tokens in plaintext. Every...

Source: DEV Community

TutorialsOrdinary2026-09-06 12:31

Malicious MCP Servers and the Unsolved Supply Chain Problem for AI Agents

On March 24, 2026, a poisoned PyPI package was live for 40 minutes. During those 40 minutes, every...

Source: DEV Community

TutorialsOrdinary2026-09-06 12:30

Agent Output Is an Injection Vector: SQL, Shell, and Template Engines

CVE-2026-26030 is not a prompt injection vulnerability. The attacker never manipulated a system...

Source: DEV Community

TutorialsOrdinary2026-09-06 11:23

Replacing Myself With AI, One Cognitive Habit at a Time

I built an AI memory system that stores what I know. The layer I actually want stores how I get curious. Then I realized what that really is: a slow machine for deleting parts of my own thinking I no longer need to do by hand. Which is als…

Source: DEV Community

TutorialsOrdinary2026-09-06 11:08

NVIDIA Just Bought the Town Square of Open-Source AI for $12.9 Billion. Here Is What You Should Actually Do About It

On September 3, NVIDIA confirmed it is buying Hugging Face for $12.93 billion. The platform that...

Source: DEV Community · NVIDIA / Hugging Face

TutorialsOrdinary2026-09-06 10:53

HypeBESTIE: My Free Sincere Compliment Engine

This is a submission for Weekend Challenge: Generosity Edition What I Built Sup Dev.to! I...

Source: DEV Community

TutorialsOrdinary2026-09-06 10:50

Machines Can Only Build What Someone Already Imagined

When a prompt describing what you need is submitted, an AI fills in forty lines of clean, working...

Source: DEV Community

TutorialsOrdinary2026-09-06 10:20

What breaks when you ship 21 AI tools that never touch a server

We run every AI feature — LLM writing, background removal, upscaling, transcription, OCR — entirely in the browser. Here is the honest list of what broke, and the COOP/COEP, onnxruntime and model-size traps we hit on the way.

Source: DEV Community

TutorialsOrdinary2026-09-06 10:17

The 404 only we could see: 23.8 hours inside a cache entry we made ourselves

For 23.8 hours, one of our published articles returned 404 to us and 200 to everyone else. The...

Source: DEV Community

TutorialsOrdinary2026-09-06 09:58

Your Hook Is Configured Correctly and Never Runs

You write a hook. The JSON is valid. The keys are spelled right. No error appears. Nothing...

Source: DEV Community

TutorialsOrdinary2026-09-06 09:47

Claims, Not Facts: Building an Auditable Multi-Author Record for a House

Most systems that describe a building store facts. A row says the house has three bedrooms. Another...

Source: DEV Community

TutorialsOrdinary2026-09-06 09:32

984 Requests Said They Were Perplexity. None Could Prove It.

Thirty days of AI crawler data from one WordPress site, sorted by whether each request could be confirmed — about one in six arrived with proof attached, and the reasons the rest could not split three ways.

Source: DEV Community

TutorialsOrdinary2026-09-06 09:30

Building Beyond CRUD with Claude and Codex: How we Created an Interactive Digital Museum

What does building with AI look like when your application needs interactive planets, a working 3D...

Source: DEV Community · Claude

TutorialsOrdinary2026-09-06 09:22

GPT-6 Astra: What’s Actually New in OpenAI’s New Frontier Model

OpenAI has released GPT-6 Astra, its newest frontier model, less than a week after Anthropic’s Claude...

Source: DEV Community · OpenAI / GPT / Claude

TutorialsOrdinary2026-09-06 09:14

handoff: Give the Next AI Agent the Context It Actually Needs

A detailed guide to creating compact AI-agent handoffs that preserve the live thread, point to durable artifacts, and prevent costly rediscovery.

Source: DEV Community

TutorialsOrdinary2026-09-06 08:48

Understanding Things Anyway

I’ve been noticing something weird lately. Not with AI itself. AI is obviously getting better: the...

Source: DEV Community

TutorialsOrdinary2026-09-06 08:44

Amparo: applying for food aid without reading a single word

This is a submission for Weekend Challenge: Generosity Edition What I Built Every year an...

Source: DEV Community

TutorialsOrdinary2026-09-06 07:54

Advertised but never wired: config options that exist everywhere except the code path

Two issues landed in the agent-CLI ecosystem within a few hours of each other this week, in two...

Source: DEV Community

TutorialsOrdinary2026-09-06 07:45

13 repositories, 13 bugs: what open source taught me about my own tool

I built a tool that draws architecture diagrams from a repository, where every edge cites the file,...

Source: DEV Community

TutorialsOrdinary2026-09-06 07:44

I built a QA agent that audits itself nightly. It has filed 79 findings against its own releases.

Most "AI QA agents" audit your repo from scratch every time and end with LGTM. I wanted the opposite:...

Source: DEV Community

TutorialsOrdinary2026-09-06 07:43

RAG ou MCP? A pergunta errada. Cole o prompt e monte os dois

RAG é a base de conhecimento: Logseq, Notion, Confluence. MCP é o conector: o IDE pede um trecho dessa base. Fine-tuning não grava o doc nos pesos.

Source: DEV Community

TutorialsOrdinary2026-09-06 07:23

urning emailed spreadsheets into Magento orders without asking customers to change

Every B2B Magento store I've worked on has the same quiet leak. A chunk of the orders never touch the...

Source: DEV Community

TutorialsOrdinary2026-09-06 07:08

Give Your Coding Agent Eyes

Every coding-agent session starts blind: your agent rediscovers how your features work, and sometimes...

Source: DEV Community

TutorialsOrdinary2026-09-06 07:01

Multiple Browser Agents Need More Than Separate Profiles

A development report on browser workspaces, project bindings, exclusive leases, and controlled handovers between humans and agents.

Source: DEV Community

TutorialsOrdinary2026-09-06 05:21

Liar Liar Pants on Fire

I have to come clean. Speaking at APIWorld this past week wasn’t actually my first talk acceptance....

Source: DEV Community

TutorialsOrdinary2026-09-06 05:11

The queue drains itself now, and the morning note fits in a minute

One month after the queue essay: findings file themselves, rounds are judged on the contract, the morning brief reads like a PA's note, and a refused write is never silent.

Source: DEV Community

TutorialsOrdinary2026-09-06 05:11

My agents run without permission prompts, so the brake moved into the hook

The permission prompt only fired when I was watching. A hook now refuses force pushes, shared-tree resets and root deletes in every mode, and redirects the session instead of waiting for me.

Source: DEV Community