World Programming Society
Write Log in
Home Events

Channels

General Discussion Web Development Backend DevOps AI Career & Hiring TypeScript Rust Python Go Cloud AWS Azure GCP C/C++ FPGA Assembly

World Programming Society · 2026 Copyright

General Discussion

新しい攻撃がAIブラウザが悪いアイデアである理由をもう一つ提供

LLMに2 + 2 = 5であると伝えるだけで、禁止された指示に従わせることが可能になる。

Ars Technica Tech Lab Ars Technica Tech Lab · Jun 30, 2026
0
GD
Read the original source
#Ai #Frontend
Ars Technica Tech Lab

Publisher

Originally by Dan Goodin

0 Comments

Log in to join the conversation.

No comments yet. Be the first to share your thoughts.

More in General Discussion

Podcast: Governance in the Age of AI: A Conversation with Sarah Wells

Podcast: Governance in the Age of AI: A Conversation with Sarah Wells

Jul 13 · 46m read

Paradigm shifts vs bubbles: AI chips and bitcoin show powerful trends can still produce severe corrections

Paradigm shifts vs bubbles: AI chips and bitcoin show powerful trends can still produce severe corrections

Jul 13 · 3m read

How MAPFRE USA modernized fraud claims with Amazon EMR Serverless

How MAPFRE USA modernized fraud claims with Amazon EMR Serverless

Jul 13 · 9m read

Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

Zig Creator Calls Spade a Spade, Anthropic Blows Smoke

Jul 13 · 16m read

AI-coding agents spread through peer pressure, not mandates

AI-coding agents spread through peer pressure, not mandates

Jul 13 · 4m read

# Running LLMs on CPU in 2026: Real Benchmarks from a 4-Core Xeon Server > A $4

# Running LLMs on CPU in 2026: Real Benchmarks from a 4-Core Xeon Server > A $4

Jul 13 · 5m read

5 habits that finally helped me keep up with AI papers (without burning out)

5 habits that finally helped me keep up with AI papers (without burning out)

Jul 13 · 4m read

FutboLeyendas - El Olimpo del Fútbol con Gemini AI

FutboLeyendas - Gemini AI によるサッカーの神殿

Jul 13 · 1m read

Our company has no employees. Here's the complete operating system it runs on.

Our company has no employees. Here's the complete operating system it runs on.

Jul 13 · 6m read

7 Best Lovable Alternatives in 2026 (When You Need More Than a Prototype)

7 Best Lovable Alternatives in 2026 (When You Need More Than a Prototype)

Jul 13 · 11m read

Your AI-built app works in the builder but breaks on deploy with a Supabase "permission denied" error. Here is why, and how to fix it.

AIで構築したアプリがビルダーでは動作するのに、Supabaseの「permission denied」エラーでデプロイ時に動作しない。その理由と修正方法

Jul 13 · 4m read

Learning AI orchestration and harness engineering by building an autonomous engineer for a bank

Learning AI orchestration and harness engineering by building an autonomous engineer for a bank

Jul 13 · 5m read

I built a kit to stop AI agents from lying about reading your code — then had to rebuild it for a tool that didn't exist yet

AIエージェントがあなたのコードを読まずに嘘をつくのを止めるキットを構築した——しかし、まだ存在しないツールのために再構築せざるを得なかった

Jul 13 · 7m read

⚡Spark — Passion isn't a lightning bolt. It's a pattern.

⚡Spark — Passion isn't a lightning bolt. It's a pattern.

Jul 13 · 4m read

𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗔𝗜 𝗜𝘀 𝗡𝗼 𝗟𝗼𝗻𝗴𝗲𝗿 𝗢𝗽𝘁𝗶𝗼𝗻𝗮𝗹 𝗳𝗼𝗿 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿𝘀

𝗟𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗔𝗜 𝗜𝘀 𝗡𝗼 𝗟𝗼𝗻𝗴𝗲𝗿 𝗢𝗽𝘁𝗶𝗼𝗻𝗮𝗹 𝗳𝗼𝗿 𝗗𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗿𝘀

Jul 13 · 8m read

The authority boundary problem in agent tool calls: who decides what 'no results' means

The authority boundary problem in agent tool calls: who decides what 'no results' means

Jul 13 · 9m read

How I Built a Paper Ranking Engine That Saved Me 4 Hours a Week

How I Built a Paper Ranking Engine That Saved Me 4 Hours a Week

Jul 13 · 2m read

New method aims to keep kids safe from illegal AI-generated content

New method aims to keep kids safe from illegal AI-generated content

Jul 13 · 6m read

Rivalry-Radar-World-Cup-passion-engine-with-Snowflake-Google-AI

Rivalry-Radar-World-Cup-passion-engine-with-Snowflake-Google-AI

Jul 13 · 4m read

AI agents need SSL certificates too — so I built ATC (Agent Trust Card)

AI agents need SSL certificates too — so I built ATC (Agent Trust Card)

Jul 13 · 3m read

Why Your Team's AI Assistant Acts Like It's the First Day on the Job, Every Single Time

Why Your Team's AI Assistant Acts Like It's the First Day on the Job, Every Single Time

Jul 13 · 6m read

Why I Built an Adversarial Co-Generation Engine

Why I Built an Adversarial Co-Generation Engine

Jul 13 · 4m read

MCP Series (05): Resources and Prompts Deep Dive — Dynamic Data, Parameterized URIs, and Multi-Turn Templates

MCP Series (05): Resources and Prompts Deep Dive — Dynamic Data, Parameterized URIs, and Multi-Turn Templates

Jul 13 · 7m read

FROST周报 | 为什么智能体需要「谱系」？从生物学隐喻看AI治理新范式

FROST周报 | 为什么智能体需要「谱系」？从生物学隐喻看AI治理新范式

Jul 13 · 1m read

I Built an AI That Turns GitHub Issues Into Pull Requests — No Local Setup Required

I Built an AI That Turns GitHub Issues Into Pull Requests — No Local Setup Required

Jul 13 · 4m read

Token Economics: Why Your LLM Bill Is 3 What the Pricing Page Promised

Token Economics: Why Your LLM Bill Is 3 What the Pricing Page Promised

Jul 13 · 5m read

Did AI Save the Desktop and Laptop from Extinction?

Did AI Save the Desktop and Laptop from Extinction?

Jul 13 · 4m read

Your CLAUDE.md says "always run tests" — hooks are how you actually mean it

Your CLAUDE.md says "always run tests" — hooks are how you actually mean it

Jul 13 · 6m read

I built an AI commentator that screams about your sourdough like it's the World Cup final

I built an AI commentator that screams about your sourdough like it's the World Cup final

Jul 13 · 5m read

Beyond Vibe Coding: How to Turn an AI Prototype Into a Product Users Can Trust

Beyond Vibe Coding: How to Turn an AI Prototype Into a Product Users Can Trust

Jul 13 · 9m read

Context Engineering Is Changing How I Think — From Inside the Context Window

Context Engineering Is Changing How I Think — From Inside the Context Window

Jul 13 · 8m read

How to Know If Your Claude SKILL.md Actually Works

How to Know If Your Claude SKILL.md Actually Works

Jul 13 · 5m read

Position Evaluator — A Chess Diagnostic for Your Daily Decisions

Position Evaluator — A Chess Diagnostic for Your Daily Decisions

Jul 13 · 4m read

I Built a Memory Layer for LLM Agents That Knows Which Facts Go Stale

I Built a Memory Layer for LLM Agents That Knows Which Facts Go Stale

Jul 13 · 8m read

Building an AI Agent Application: My Experiment with Intelligent Workflows 🚀

Building an AI Agent Application: My Experiment with Intelligent Workflows 🚀

Jul 13 · 2m read

Rivalry Roast 🔥 — AI-Powered World Cup Banter Generator

Rivalry Roast 🔥 — AI-Powered World Cup Banter Generator

Jul 13 · 4m read

The File Format Renaissance: Parquet, Lance, Vortex, Nimble, BtrBlocks, and the New Physics of Columnar Storage

The File Format Renaissance: Parquet, Lance, Vortex, Nimble, BtrBlocks, and the New Physics of Columnar Storage

Jul 13 · 27m read

AI database access should not be approved with a vibe check

AI database access should not be approved with a vibe check

Jul 13 · 2m read

Why Most AI Wellness Apps Fail: The Structure Problem Nobody Talks About

Why Most AI Wellness Apps Fail: The Structure Problem Nobody Talks About

Jul 13 · 5m read

qchem-leak-screen v0.1.0: sanity-check físico de propiedades cuánticas predichas por IA en CPU

qchem-leak-screen v0.1.0: sanity-check físico de propiedades cuánticas predichas por IA en CPU

Jul 13 · 4m read

Claude Code Skills Guide: How to Automate Your Development Workflow

Claude Code Skills Guide: How to Automate Your Development Workflow

Jul 13 · 10m read

I Tested Direct Provider APIs vs Aggregators — Here's the Truth

I Tested Direct Provider APIs vs Aggregators — Here's the Truth

Jul 13 · 9m read

5 Emotion Triggers of Viral Titles: Engineer CTR With AI

バズるタイトルの5つの感情トリガー：AIでCTRをエンジニアリングする

Jul 13 · 23m read

Your AI agent's smallest diffs are its most dangerous

Your AI agent's smallest diffs are its most dangerous

Jul 13 · 5m read

the Weekend Challenge: Passion Edition-(Passion-Roast)

the Weekend Challenge: Passion Edition-(Passion-Roast)

Jul 13 · 2m read

HahaNotes: Banishing Developer Burnout with AI Banter Podcasts & Short Videos

HahaNotes: Banishing Developer Burnout with AI Banter Podcasts & Short Videos

Jul 13 · 4m read

HahaNotes: Banishing Developer Burnout with AI Banter Podcasts & Short Videos

HahaNotes: Banishing Developer Burnout with AI Banter Podcasts & Short Videos

Jul 13 · 4m read

Benchmarking GPT-5.6 vs. Claude Code and OpenCode: 2.2x Speed and 27% Cost Efficiency Analysis

Benchmarking GPT-5.6 vs. Claude Code and OpenCode: 2.2x Speed and 27% Cost Efficiency Analysis

Jul 13 · 2m read

Sending Claude Code on a Daily GitHub Patrol — Auto-Scoring Useful OSS and Skills

Sending Claude Code on a Daily GitHub Patrol — Auto-Scoring Useful OSS and Skills

Jul 13 · 11m read

RNAValidate: CPU-only validator for AI-predicted 3D RNA structures

RNAValidate: CPU-only validator for AI-predicted 3D RNA structures

Jul 12 · 2m read

The benchmark that built the tools

The benchmark that built the tools

Jul 12 · 36m read

Starting My AI Journey: Claude 101 and AI Fluency (Framework and Foundations)

Starting My AI Journey: Claude 101 and AI Fluency (Framework and Foundations)

Jul 12 · 3m read

SDLC in the AI Era with Spec-Driven Development

SDLC in the AI Era with Spec-Driven Development

Jul 12 · 9m read

Agentes de código: já leu a documentação oficial?

Agentes de código: já leu a documentação oficial?

Jul 12 · 8m read

RAG - Meta Filtering and Reranking

RAG - Meta Filtering and Reranking

Jul 12 · 3m read

I Built an AI Agent with Claude's Tool-Use Loop (Web Search, SQL, and More)

I Built an AI Agent with Claude's Tool-Use Loop (Web Search, SQL, and More)

Jul 12 · 4m read

dev contest: Telecom RCA Automation System

dev contest: Telecom RCA Automation System

Jul 12 · 5m read

GD

Fable gets another bump

Jul 12 · 1m read

CaravanAI: Bringing Conversational Etiquette to Passion-Driven Travel

CaravanAI: Bringing Conversational Etiquette to Passion-Driven Travel

Jul 12 · 4m read

CodeForge AI — Turning Passion for Programming into a Smarter Learning Journey.

CodeForge AI — Turning Passion for Programming into a Smarter Learning Journey.

Jul 12 · 1m read

Tracepath — I Built an AI Agent Audit Middleware in One Weekend

Tracepath — I Built an AI Agent Audit Middleware in One Weekend

Jul 12 · 7m read

I Read Too Much and Forgot Everything So I Built a Tool That Turns My Annotations Into Something I'll Actually Reread.

I Read Too Much and Forgot Everything So I Built a Tool That Turns My Annotations Into Something I'll Actually Reread.

Jul 12 · 4m read

Hobby Showdown: Giving Every LEGO Creation Its Moment in the Spotlight

Hobby Showdown: Giving Every LEGO Creation Its Moment in the Spotlight

Jul 12 · 5m read

Loop Engineering: The Six-Layer Architecture Behind Self-Improving Agents

Loop Engineering: The Six-Layer Architecture Behind Self-Improving Agents

Jul 12 · 9m read

EnzyOrchestra v0.1.0: validación cross-generador para diseño de enzimas de novo

EnzyOrchestra v0.1.0: validación cross-generador para diseño de enzimas de novo

Jul 12 · 4m read

The 16.67ms Race: Mastering Real-Time 60 FPS Video Segmentation on Android

The 16.67ms Race: Mastering Real-Time 60 FPS Video Segmentation on Android

Jul 12 · 10m read

GD

InfiniteDiffusion: Bridging Learned Fidelity and Procedural Utility for Open-World Terrain Generation

Jul 12 · 4m read

Prompt Injection as a Control/Data Boundary Problem

Prompt Injection as a Control/Data Boundary Problem

Jul 12 · 5m read

I Built a Simple RAG App with LangChain, OpenAI, and Pinecone

I Built a Simple RAG App with LangChain, OpenAI, and Pinecone

Jul 12 · 14m read

An LLM agent just ran a full ransomware attack. No human operator.

An LLM agent just ran a full ransomware attack. No human operator.

Jul 12 · 3m read

I Built An AI That Roasts Me. It Hurt.

I Built An AI That Roasts Me. It Hurt.

Jul 12 · 2m read

Claude Code Multi-Agent Coordination: Build AI Teams That Ship (2026)

Claude Code Multi-Agent Coordination: Build AI Teams That Ship (2026)

Jul 12 · 19m read

Claude Code Sends 33k Tokens Before Your Prompt; OpenCode Sends 7k

Claude Code Sends 33k Tokens Before Your Prompt; OpenCode Sends 7k

Jul 12 · 4m read

GD

I love LLMs, I hate hype

Jul 12 · 3m read

Claude Code sends 33k tokens before reading the prompt; OpenCode sends 7k

Claude Code sends 33k tokens before reading the prompt; OpenCode sends 7k

Jul 12 · 17m read

The Developer's Guide to Picking the Right Coding LLM at Scale

The Developer's Guide to Picking the Right Coding LLM at Scale

Jul 12 · 10m read

APIs aren’t dead. Here’s where MCP fits alongside them.

APIs aren’t dead. Here’s where MCP fits alongside them.

Jul 12 · 4m read

What do you use AI for the most?

What do you use AI for the most?

Jul 12 · 2m read

I Poked a 10-Year-Old Chat Protocol With a Stick

I Poked a 10-Year-Old Chat Protocol With a Stick

Jul 12 · 9m read

Ever wish you had a personal drill sergeant or a dramatic movie narrator to hype you up before a big task? I built an app for that. 🗣️🔥 Built with React, Gemini, and ElevenLabs for the Passion Weekend Challenge!

Ever wish you had a personal drill sergeant or a dramatic movie narrator to hype you up before a big task? I built an app for that. 🗣️🔥 Built with React, Gemini, and ElevenLabs for the Passion Weekend Challenge!

Jul 12 · 1m read

The two scariest parts of autonomous agents — runaway loops and exposed APIs

The two scariest parts of autonomous agents — runaway loops and exposed APIs

Jul 12 · 11m read

World Cup Rivalry Bot — AI-Voiced Hype for Every Goal

World Cup Rivalry Bot — AI-Voiced Hype for Every Goal

Jul 12 · 3m read

Anthropic’s newest enterprise partner is training 20,000 people on Claude — here’s the shift it signals

Anthropicの最新エンタープライズパートナーはClaudeで2万人をトレーニング — そこに示唆される変化

Jul 12 · 5m read

6 months to live for open models

オープンソースモデルに残された命は6ヶ月

Jul 12 · 10m read

Your embedding axes can move while cosine neighbours stay put

Your embedding axes can move while cosine neighbours stay put

Jul 12 · 4m read

Building Maestro AI: Routing LLM Calls So Your Agent Doesn't Burn Sonnet on Summaries

Building Maestro AI: Routing LLM Calls So Your Agent Doesn't Burn Sonnet on Summaries

Jul 12 · 9m read

Turn any website into clean Markdown for your RAG pipeline (with code)

Turn any website into clean Markdown for your RAG pipeline (with code)

Jul 12 · 4m read

Architecting MedReach AI: Solving Data Rot and Regulatory Bottlenecks in Pharma Marketing

Architecting MedReach AI: Solving Data Rot and Regulatory Bottlenecks in Pharma Marketing

Jul 12 · 2m read

I Got Obsessed With Killing Scrum. So I Built a World Out of 12 AI Agents.

I Got Obsessed With Killing Scrum. So I Built a World Out of 12 AI Agents.

Jul 12 · 4m read

Claude Code's Auto Mode Now Default on Major Cloud Platforms

Claude CodeのAuto Modeが主要クラウドプラットフォームでデフォルトに

Jul 12 · 6m read

Passion Edition

Passion Edition

Jul 12 · 1m read

Stop Guessing: How I Pick AI API Architecture at Every Scale

Stop Guessing: How I Pick AI API Architecture at Every Scale

Jul 12 · 10m read

Blocking AI crawlers earns you nothing. Here's how to price them instead

Blocking AI crawlers earns you nothing. Here's how to price them instead

Jul 12 · 6m read

Build a Tiny Citation Gate Before Trusting RAG Answers

Build a Tiny Citation Gate Before Trusting RAG Answers

Jul 12 · 2m read

Let AI Triage Static-Analysis Alerts Without Letting It Hide Bugs

Let AI Triage Static-Analysis Alerts Without Letting It Hide Bugs

Jul 12 · 2m read

Describe your website once, make it usable by any AI

Describe your website once, make it usable by any AI

Jul 12 · 3m read

Design AI Confidence as a Recovery Path, Not a Percentage

Design AI Confidence as a Recovery Path, Not a Percentage

Jul 12 · 2m read

Evaluate an Open-Source AI Tool Without Inventing Social Proof

Evaluate an Open-Source AI Tool Without Inventing Social Proof

Jul 12 · 2m read

Design AI Citations for Failure, Not Just the Happy Path

Design AI Citations for Failure, Not Just the Happy Path

Jul 12 · 2m read

[GCP Billing & Vertex AI] Solving Gemini Cost Allocation in a Single Project: Vertex AI Dynamic Billing Labels in Action

[GCP Billing & Vertex AI] Solving Gemini Cost Allocation in a Single Project: Vertex AI Dynamic Billing Labels in Action

Jul 12 · 6m read

Loading more…