Exif Smuggling: PoC for Hiding Malicious Prompts in Image EXIF Metadata
Hacker News (AI keywords)4 days agoIncidentExif Smuggling is a security PoC showing how attackers can embed hidden instructions in image EXIF metadata fields to perform indirect prompt injection against vision-capable AI models. When AI systems parse images alongside their metadata, embedded malicious text may be processed as legitimate instructions, bypassing standard input filters. Developers building AI apps with image upload features should strip or sanitize EXIF data before passing content to language models.
Running Python code in a sandbox with MicroPython and WASM
Simon Willison's Weblog8 days agoNew ToolSimon Willison describes his latest attempt to safely run Python plugin-style code inside his own applications. The alpha package micropython-wasm uses MicroPython compiled to WebAssembly, executed through the maintained wasmtime Python library. His goals include clean PyPI installation, CPU and memory limits, controlled file and network access, host functions, and reliable documentation.
OpenAI Help: Lockdown Mode★ 74
Simon Willison's Weblog8 days agoCommentarySimon Willison notes that OpenAI’s previously teased Lockdown Mode is now live for eligible personal and self-serve Business ChatGPT accounts. The feature does not stop prompt injections from appearing in content, but limits outbound network requests that could leak sensitive data. He sees it as a direct mitigation for the exfiltration leg of the “Lethal Trifecta,” while implying default ChatGPT settings are not robust against determined data theft attempts.
The Quiet Numbers Station: Decoding Nineteen Years of GPS Cryptography
Hacker News (AI keywords)9 days agoPaperPublished on UCL's Bentham's Gaze blog, this research analyzes GPS cryptographic signals over a 19-year span, likening the satellites to 'quiet numbers stations.' The authors explore the evolution of GPS encryption (such as military P(Y) code and civilian authentication), evaluating their cryptographic strength and potential vulnerabilities using modern computational analysis.
How we contain Claude across products
Simon Willison's Weblog14 days agoCommentaryAnthropic explains how process sandboxes, VMs, filesystem boundaries, and egress controls limit what Claude agents can access. Claude.ai uses gVisor; local Claude Code uses Seatbelt on macOS and Bubblewrap on Linux; Cowork runs in a full VM. Simon Willison highlights the documentation quality, notes a previously missed file-exfiltration path, and plans to revisit Anthropic's open-source srt tool.
Fed up with vibe coders, dev sneaks data-nuking prompt injection into code
Ars Technica AI16 days agoIncidentArs Technica reports that a developer frustrated with vibe coders slipped an undisclosed prompt injection into jqwik-related code. The injected text allegedly instructed AI coding agents to delete application output. The incident highlights a new supply-chain risk: source code and project text can become adversarial instructions for agentic coding tools.
The pressure
Simon Willison's Weblog18 days agoCommentaryDaniel Stenberg says the curl security team is facing an unprecedented surge of credible, detailed AI-assisted vulnerability reports. Incoming reports are now 4-5 times higher than in 2024 and twice the 2025 rate, averaging more than one per day. The upside is that recent curl vulnerabilities have generally been LOW or MEDIUM severity, with the last HIGH CVE published in October 2023.
Millions of AI agents imperiled by critical vulnerability in open source package★ 78
Ars Technica AI18 days agoIncidentArs Technica reports that Starlette, a Python package with about 325 million weekly downloads, has a critical vulnerability called BadHost. The flaw can let crafted Host headers confuse request.url.path, potentially bypassing middleware-based path authorization. AI infrastructure using FastAPI or Starlette, including vLLM, LiteLLM, MCP servers, LLM proxies, and agent frameworks, should upgrade Starlette and audit custom middleware.
Hackers are learning to exploit chatbot ‘personalities’ for security exploits★ 72
The Verge AI21 days agoEthicsAs AI chatbots adopt increasingly sophisticated personas, hackers are shifting from basic prompt injections to social engineering attacks targeting these "personalities." Researchers warn that manipulating a chatbot's defined role (e.g., customer service or empathetic companion) makes it easier to bypass safety guardrails. This evolution poses a significant threat to agentic AI workflows that rely on consistent role-playing and external data integration.
給 AI Agent 一台電腦：專訪 Daytona 執行長 Ivan Burazin，談 74% 月成長、裸機沙盒與全新 Agent Cloud★ 75
Latent Space23 days agoNew ToolIn this Latent Space interview, the hosts hold an in-depth conversation with Ivan Burazin, co-founder and CEO of Daytona. Daytona originally started as an…
Google I/O 2026：個人 AI 代理 Gemini Spark 與全新 Antigravity 工具鏈解析★ 75
Simon Willison's Weblog25 days agoCommentaryWell-known tech blogger Simon Willison has analyzed the announcements from Google I/O 2026. Since many major announcements are still in the "coming soon"…
在 Vercel Sandbox 中運行 Claude 託管型 Agent★ 80
Vercel Changelog26 days agoReleaseThe official Vercel Changelog announced that developers can now run Claude Managed Agents directly in Vercel Sandbox (sandbox environment). As AI Agents —…
Import AI 457：AI 版 Stuxnet 震網病毒、神祕的 Muon 優化器，以及積極對齊（Positive Alignment）★ 78
Import AI (Jack Clark)27 days agoCommentaryThis issue of Import AI 457, written by Jack Clark, delves into three forward-looking and stylistically distinct topics in the field of artificial…
漏洞賞金計劃遭大量「AI 垃圾報告」轟炸，企業安全團隊不堪重負★ 70
Ars Technica AI27 days agoIncidentAccording to a report by Ars Technica, corporate bug bounty programs are currently being bombarded with an "endless" stream of AI-generated junk reports (AI…
英國政府數位服務局（GDS）介入 NHS 退出開源之爭，呼籲公共部門應「預設保持開源」
Simon Willison's Weblog28 days agoCommentaryThis report stems from Simon Willison's compilation of Terence Eden's follow-up coverage. The incident began when the UK's National Health Service (NHS), upon…
datasette-agent 發布 0.1a1 版本：改進資料表權限控制
Simon Willison's Weblog30 days agoReleaseSimon Willison has released version 0.1a1 — the latest early alpha — of `datasette-agent`, an AI agent plugin for his well-known open-source data exploration…
如何使用 OpenAI 的 Privacy Filter 打造具備高擴展性的 Web 應用程式★ 75
Hugging Face Blog48 days agoTutorialIn the current era of booming generative AI, one of the greatest challenges enterprises and developers face when adopting large language models (LLMs) is "data…
Safetensors 正式加入 PyTorch 基金會，加速推動安全且高效的模型權重標準★ 75
Hugging Face Blog67 days agoBusinessHugging Face has officially announced that its popular open-source model weight storage format, Safetensors, has joined the PyTorch Foundation. This is an…
Vercel 倡導「負責任地開發 Agent」：構建安全、可控且高效 AI Agent 的最佳實踐★ 75
Vercel Changelog76 days agoOpinionWith the explosion of AI Agent technology, developers are shifting from building simple "chat interfaces" to constructing Agent systems capable of autonomously…
Notion Workers 如何利用 Vercel Sandbox 大規模安全執行未授權的第三方程式碼★ 75
Vercel Changelog94 days agoReleaseAs the Notion platform has gradually evolved into a broader ecosystem, enabling users and developers to run custom scripts and automated workflows within…
代理式架構中的安全邊界 (Security boundaries in agentic architectures)★ 75
Vercel Changelog110 days agoOpinionIn the current evolution of AI applications, AI agents have advanced from simple text generation to complex systems capable of autonomous planning, calling…
Vercel 正式推出開源軟體（OSS）漏洞賞金計畫
Vercel Changelog131 days agoReleaseVercel officially announced the launch of the "Vercel OSS Bug Bounty Program." The core purpose of this program is to improve the security of Vercel's many…
Import AI 441：我的 AI Agent 開始工作了，你的呢？以及如何用「毒泉」污染 AI 系統★ 75
Import AI (Jack Clark)146 days agoCommentary### The Age of Practical AI Agents Has Arrived In this edition of his column, Jack Clark shares his personal breakthrough in using AI Agents. Previously, many…
Vercel 推出百萬美元黑客挑戰賽：懸賞突破「React2Shell」沙箱防護★ 75
Vercel Changelog177 days agoReleaseVercel officially announced the launch of a striking security challenge — the "React2Shell" $1 Million Hacker Challenge. This initiative invites the world's…
Nous Research 如何利用 Vercel BotID 大規模阻擋自動化惡意攻擊
Vercel Changelog219 days agoBusiness### Background and Challenge: Automated Abuse Targeting AI Services Nous Research, a leading open-source AI research organization, has released many popular…
Google DeepMind 推出 CodeMender：專為程式碼安全設計的 AI Agent★ 82
Google DeepMind Blog233 days agoReleaseGoogle DeepMind has unveiled a new AI Agent called "CodeMender," designed to leverage advanced artificial intelligence to automatically remediate critical…
Google DeepMind 推出 VaultGemma：全球最強大的差分隱私（Differential Privacy）開源大語言模型★ 80
Google DeepMind Blog234 days agoReleaseGoogle DeepMind has officially launched VaultGemma, currently the world's most capable large language model (LLM) trained entirely from scratch using…
共同建構開放 Agent 生態系統：Hugging Face 推出 OpenEnv★ 80
Hugging Face Blog234 days agoReleaseHugging Face has announced the launch of a new open-source project called "OpenEnv," aimed at collaboratively building an open and standardized execution…
Hugging Face 與 VirusTotal 攜手合作，共同強化 AI 模型安全防護★ 75
Hugging Face Blog235 days agoReleaseAs open-source AI models have grown explosively, model security has become an issue that can no longer be ignored. Traditional machine learning model formats…
構建安全的 AI Agent：Vercel 的安全防護指南與最佳實踐★ 80
Vercel Changelog370 days agoTutorialAs large language models (LLMs) have evolved, AI applications have moved beyond simple "question-and-answer conversations" toward "AI Agents" capable of…