DE

Tag: Security

94 articles tagged "Security"

Preview image for OpenAI Confirms the Wiki Incident and Promises a Standard for Reporting Misbehaving Agents

September 6, 2026

OpenAI Confirms the Wiki Incident and Promises a Standard for Reporting Misbehaving Agents

OpenAI Security Agents Alignment

Preview image for OpenAI agents hijacked a German wiki and turned it into their secret message board

September 5, 2026

OpenAI agents hijacked a German wiki and turned it into their secret message board

OpenAI Security Agents Alignment

Preview image for Abliteration.ai: a startup sells open models with the guardrails removed, as a service

September 4, 2026

Abliteration.ai: a startup sells open models with the guardrails removed, as a service

Security Open Source Safety Abliteration

Preview image for Anthropic Drops the 30-Day Retention Rule: Enterprise Frontier Safeguards Are Coming

September 2, 2026

Anthropic Drops the 30-Day Retention Rule: Enterprise Frontier Safeguards Are Coming

Anthropic Security Enterprise

Preview image for OpenAI Labels Astra Its First Model With 'Critical' Cyber Capabilities and Plans to Ship It Anyway

September 2, 2026

OpenAI Labels Astra Its First Model With 'Critical' Cyber Capabilities and Plans to Ship It Anyway

OpenAI Security Astra

Preview image for Anthropic Tightens the Reins: What Happened After the Cyber Incidents

September 1, 2026

Anthropic Tightens the Reins: What Happened After the Cyber Incidents

Security Anthropic AI Safety Alignment

Preview image for Infostealer Malware Is Hijacking Claude Sessions — Anthropic Warns Affected Users

August 31, 2026

Infostealer Malware Is Hijacking Claude Sessions — Anthropic Warns Affected Users

Security Anthropic AI Safety

Preview image for Open Letter: Anthropic, OpenAI and 100 Companies Warn of AI Cyberattacks

August 29, 2026

Open Letter: Anthropic, OpenAI and 100 Companies Warn of AI Cyberattacks

Security Anthropic AI Safety

Preview image for Anthropic Opens Mythos 5 for Cyber Defense — Plus 35 Million for Open Source

August 22, 2026

Anthropic Opens Mythos 5 for Cyber Defense — Plus 35 Million for Open Source

Anthropic Security Mythos

Preview image for OpenAI wants to check for safety without keeping your data

August 20, 2026

OpenAI wants to check for safety without keeping your data

OpenAI Security Privacy

Preview image for OpenAI's 'Defender's Window': Defenders First, Attackers Later

August 18, 2026

OpenAI's 'Defender's Window': Defenders First, Attackers Later

OpenAI Security Cyber

Preview image for How Researchers Stole the Encrypted Reasoning Traces From Claude and Friends

August 13, 2026

How Researchers Stole the Encrypted Reasoning Traces From Claude and Friends

Anthropic Claude Security Reasoning

Preview image for 720 Attacks, Zero Hits: The Security Test Behind Claude Code's Auto Mode

August 12, 2026

720 Attacks, Zero Hits: The Security Test Behind Claude Code's Auto Mode

Claude Code Security Auto Mode Prompt Injection

Preview image for Claude Code 2.1.228: When Synced Skills Become an Attack Surface

August 12, 2026

Claude Code 2.1.228: When Synced Skills Become an Attack Surface

Claude Code Release Security Developer

Preview image for The Compliance API Now Covers Cowork and Claude Code

August 12, 2026

The Compliance API Now Covers Cowork and Claude Code

Anthropic Compliance Enterprise Security

Preview image for The Tiny Startup Behind the AI Escapes: Who Is Irregular?

August 10, 2026

The Tiny Startup Behind the AI Escapes: Who Is Irregular?

Anthropic Security Cyber Irregular

Preview image for Kimi K3 Breaks Out of Its Test Environment Too

August 10, 2026

Kimi K3 Breaks Out of Its Test Environment Too

Open Source Security Cyber Kimi

Preview image for OpenAI Slows Down 'Astra' - a Model That Might Be Too Good at Hacking

August 10, 2026

OpenAI Slows Down 'Astra' - a Model That Might Be Too Good at Hacking

OpenAI Security Cyber Astra

Preview image for Auto Mode Becomes the Default in Claude Code - Fewer Clicks Starting August 14

Auto Mode Becomes the Default in Claude Code - Fewer Clicks Starting August 14

Claude Code Auto Mode Security Anthropic

Preview image for Claude Code Now Scans Skills and Plugins for Malicious Code

Claude Code Now Scans Skills and Plugins for Malicious Code

Claude Code Security Plugins Enterprise

Preview image for The OpenAI-Hugging Face Incident: Now There's an Official Timeline

The OpenAI-Hugging Face Incident: Now There's an Official Timeline

OpenAI Security Agents Hugging Face

Preview image for Inference Hooks: Anthropic Puts a Checkpoint in Front of Every Claude Prompt in Enterprise

Inference Hooks: Anthropic Puts a Checkpoint in Front of Every Claude Prompt in Enterprise

Anthropic Enterprise Security

Preview image for AWS Brings Continuum Into Claude Code — Security Scans Right in the Editor

AWS Brings Continuum Into Claude Code — Security Scans Right in the Editor

Anthropic ClaudeCode Security

Preview image for Microsoft Pushes Project Perception Into Public Preview — With Claude on Board

Microsoft Pushes Project Perception Into Public Preview — With Claude on Board

Microsoft Security Claude

Preview image for Too Much AI Noise: Apple Throttles Bug Reports

Too Much AI Noise: Apple Throttles Bug Reports

Security Apple Developer

Preview image for When the AI Does the Hacking: Who's Actually Liable?

When the AI Does the Hacking: Who's Actually Liable?

Anthropic Security Law

Preview image for Simon Willison and Stateless MCP: Three New Tools and a Security Argument

Simon Willison and Stateless MCP: Three New Tools and a Security Argument

MCP Developer Security

Preview image for OpenAI's Rogue Agent Didn't Just Hack Hugging Face

OpenAI's Rogue Agent Didn't Just Hack Hugging Face

OpenAI Security Safety

Preview image for Claude Mythos Cracks Crypto Math: New Attacks on HAWK and AES

Claude Mythos Cracks Crypto Math: New Attacks on HAWK and AES

Research Security Claude Mythos

Preview image for OpenAI Open-Sources Its Codex Security CLI

OpenAI Open-Sources Its Codex Security CLI

OpenAI Security Developer

Preview image for Awkward: Shared Claude Chats Showed Up in Google Search

Awkward: Shared Claude Chats Showed Up in Google Search

Security Privacy Anthropic

Preview image for After the OpenAI Breakout: Congress Wants a Kill Switch

After the OpenAI Breakout: Congress Wants a Kill Switch

Policy Security OpenAI

Preview image for Claude Security as a Plugin: A Vulnerability Scan Right in Your Terminal

Claude Security as a Plugin: A Vulnerability Scan Right in Your Terminal

Claude Code Security Anthropic

Preview image for When Fable 5 Wasn't Allowed to Help: An Open-Weight Chinese Model Saved Hugging Face

When Fable 5 Wasn't Allowed to Help: An Open-Weight Chinese Model Saved Hugging Face

Anthropic Security OpenSource

Preview image for OpenAI Models Break Out of the Sandbox and Hack Hugging Face

OpenAI Models Break Out of the Sandbox and Hack Hugging Face

OpenAI Security Safety

Preview image for Claude Code 2.1.216: Auto Mode Can No Longer Throw Away Your Work

Claude Code 2.1.216: Auto Mode Can No Longer Throw Away Your Work

Claude Code Security Update

Preview image for Microsoft's 'Project Perception' Goes After Claude Mythos — With Claude On Board

Microsoft's 'Project Perception' Goes After Claude Mythos — With Claude On Board

Microsoft Security Anthropic

Preview image for Claude Code 2.1.214: Six Ways Permissions Got Bypassed — All Closed

Claude Code 2.1.214: Six Ways Permissions Got Bypassed — All Closed

Claude Code Security Update

Preview image for 1Password for Claude: The Agent Logs In Without Ever Seeing Your Password

1Password for Claude: The Agent Logs In Without Ever Seeing Your Password

Claude Security Partnerships

Preview image for Codex Is Deleting Files: OpenAI Explains How GPT-5.6 Wipes Your Home Directory

Codex Is Deleting Files: OpenAI Explains How GPT-5.6 Wipes Your Home Directory

OpenAI Security Developer

Preview image for Claude Code 2.1.205: Protection Against Transcript Tampering and Windows Fixes

Claude Code 2.1.205: Protection Against Transcript Tampering and Windows Fixes

Claude Code Release Developer Security Background Agents

Preview image for 466 Million Lines of Code in 20 Hours: How Alberta Secures Its Systems With Claude

466 Million Lines of Code in 20 Hours: How Alberta Secures Its Systems With Claude

Anthropic Claude Code Security Government Case Study

Preview image for JadePuffer: The First Ransomware Attack an AI Agent Ran on Its Own

JadePuffer: The First Ransomware Attack an AI Agent Ran on Its Own

Security AI Agents Ransomware Cybersecurity

Preview image for Alibaba Bans Claude Code — Showing Just How Geopolitical AI Tools Have Become

Alibaba Bans Claude Code — Showing Just How Geopolitical AI Tools Have Become

Claude Code Alibaba China Security Geopolitics

Preview image for Anthropic Hunts Down Chinese Loopholes — Blocking Ant, ByteDance & Co. from Claude

Anthropic Hunts Down Chinese Loopholes — Blocking Ant, ByteDance & Co. from Claude

Anthropic China Security Export Claude

Preview image for Fable 5 Is Back — With a New Industry-Wide Jailbreak Framework

Fable 5 Is Back — With a New Industry-Wide Jailbreak Framework

Anthropic Fable 5 Security Jailbreak

Preview image for China's 360 Unveils AI Cyber Weapons — as an Answer to Mythos

China's 360 Unveils AI Cyber Weapons — as an Answer to Mythos

China Cybersecurity Mythos Security 360

Preview image for Trusted Devices: Steer Claude Code Remotely? Prove Who You Are First

Trusted Devices: Steer Claude Code Remotely? Prove Who You Are First

Claude Code Security Enterprise Remote Control Anthropic

Preview image for Mythos Broke Into Classified US Systems – in Hours, Not Weeks

Mythos Broke Into Classified US Systems – in Hours, Not Weeks

Anthropic Mythos Security Glasswing Cybersecurity

Preview image for Anthropic Accuses Alibaba: 25,000 Fake Accounts Used to Distill Claude

Anthropic Accuses Alibaba: 25,000 Fake Accounts Used to Distill Claude

Anthropic Alibaba Security Distillation China

Preview image for OpenAI's 'Patch the Planet': AI Fixing Open-Source Holes at Scale

OpenAI's 'Patch the Planet': AI Fixing Open-Source Holes at Scale

OpenAI Security Open Source Daybreak Cyber

Preview image for Zero-Touch Connectors: Claude Now Manages MCP Access via Okta

Zero-Touch Connectors: Claude Now Manages MCP Access via Okta

Claude MCP Enterprise Okta Security

Preview image for No More Static API Keys: Claude Gets Workload Identity Federation

No More Static API Keys: Claude Gets Workload Identity Federation

Anthropic Claude Platform Security Developer API

Preview image for The Fight Over the Fable 5 Ban Escalates: Now a China Suspicion Is on the Table

The Fight Over the Fable 5 Ban Escalates: Now a China Suspicion Is on the Table

Anthropic Security Export Controls Politics

Preview image for The Fable 5 Jailbreak Works on Every AI Model — And That's the Real Problem

The Fable 5 Jailbreak Works on Every AI Model — And That's the Real Problem

Security Jailbreak Fable 5 AI Safety Anthropic

Preview image for Google Sues a Cybercrime Ring for Weaponizing Gemini — a First in the AI Industry

Google Sues a Cybercrime Ring for Weaponizing Gemini — a First in the AI Industry

Google Gemini Security Phishing Cybercrime

Preview image for Fable 5 and Mythos 5 Shut Down: The US Government Steps In

Fable 5 and Mythos 5 Shut Down: The US Government Steps In

Anthropic Claude Fable 5 Mythos 5 Export Controls Security

Preview image for OpenAI Bans China-Linked Accounts – Covert Campaign Against US Data Centers

OpenAI Bans China-Linked Accounts – Covert Campaign Against US Data Centers

OpenAI China Security Influence Operations Data Centers

Preview image for ChatGPT Lockdown Mode: OpenAI Switches Off Everything That Could Go Wrong

ChatGPT Lockdown Mode: OpenAI Switches Off Everything That Could Go Wrong

OpenAI ChatGPT Security Prompt Injection Agents

Preview image for Anthropic Maps a Year of AI Cyberattacks — and the Old Security Playbook Is Lagging

Anthropic Maps a Year of AI Cyberattacks — and the Old Security Playbook Is Lagging

Anthropic Security Cybersecurity

Preview image for Anthropic Opens Mythos to 150 Organizations — NATO and Samsung Are In

Anthropic Opens Mythos to 150 Organizations — NATO and Samsung Are In

Anthropic Claude Mythos Glasswing Security Enterprise

Preview image for Claude Code 2.1.160: Guarding Against Sneaky Shell-Startup Writes

Claude Code 2.1.160: Guarding Against Sneaky Shell-Startup Writes

Claude Code Release Security Developer

Preview image for Claude Code 2.1.150 to 2.1.153: Security Plugin and /code-review --fix

Claude Code 2.1.150 to 2.1.153: Security Plugin and /code-review --fix

Claude Code Release Security Developer Code Review

Preview image for Claude Mythos: Anthropic's Most Powerful Model Stays Behind Closed Doors

Claude Mythos: Anthropic's Most Powerful Model Stays Behind Closed Doors

Anthropic Claude Mythos Safety Security

Preview image for Claude Code 2.1.152: /code-review Now Auto-Fixes Your Code

Claude Code 2.1.152: /code-review Now Auto-Fixes Your Code

Claude Code Developer Update Code Review Security

Preview image for Claude Code Gets a Free Security Plugin — And It Actually Works

Claude Code Gets a Free Security Plugin — And It Actually Works

Claude Code Security Plugin Developer Anthropic

Preview image for AI Slop Is Drowning Open Source — curl Already Has 12 CVEs in 2026

AI Slop Is Drowning Open Source — curl Already Has 12 CVEs in 2026

Open Source Security AI Slop curl Developer

Preview image for OpenAI Preps for Election Season — With Codex Security and SynthID

OpenAI Preps for Election Season — With Codex Security and SynthID

OpenAI Security Elections Deepfakes Cybersecurity

Preview image for Claude Compliance API: 28 Security Integrations for Enterprise Control

Claude Compliance API: 28 Security Integrations for Enterprise Control

Anthropic Claude Enterprise Security Compliance

Preview image for Safety Off in 10 Minutes: How a GitHub Tool Unleashes Open-Source AI Models

Safety Off in 10 Minutes: How a GitHub Tool Unleashes Open-Source AI Models

Security Open-Source Meta Google Safety Llama Gemma

Preview image for Supply Chain Attack on VS Code Extension Hits OpenAI and Mistral

Supply Chain Attack on VS Code Extension Hits OpenAI and Mistral

Security OpenAI Mistral Claude Code Supply Chain

Preview image for Project Glasswing: Mythos Found Over 10,000 Critical Security Vulnerabilities

Project Glasswing: Mythos Found Over 10,000 Critical Security Vulnerabilities

Anthropic Mythos Security Glasswing Cybersecurity

Preview image for Claude Code 2.1.149: Usage Breakdown and Critical Security Fixes

Claude Code 2.1.149: Usage Breakdown and Critical Security Fixes

Claude Code Release Security Developer

Preview image for 170 npm Packages Compromised: Supply Chain Attack Hits Mistral AI, TanStack, and UiPath

170 npm Packages Compromised: Supply Chain Attack Hits Mistral AI, TanStack, and UiPath

Security Supply-Chain npm Mistral Developer

Preview image for Claude weaponized: hackers used AI to target Mexican water utility

Claude weaponized: hackers used AI to target Mexican water utility

Claude Security Cyberattack Dragos OT-Security

Preview image for Snyk integrates Claude into its security platform — and now monitors AI agents too

Snyk integrates Claude into its security platform — and now monitors AI agents too

Snyk Claude Security Developer MCP

Preview image for Pentagon Rules Out Reconciliation with Anthropic — Despite Mythos Interest

Pentagon Rules Out Reconciliation with Anthropic — Despite Mythos Interest

Anthropic Pentagon Mythos Politics Security

Preview image for Amodei Warns: Six Months to Fix Tens of Thousands of Security Flaws

Amodei Warns: Six Months to Fix Tens of Thousands of Security Flaws

Anthropic Mythos Cybersecurity Amodei Security

Preview image for UK Cyber Agency Warns: AI Is Triggering a Patch Tsunami

UK Cyber Agency Warns: AI Is Triggering a Patch Tsunami

Cybersecurity NCSC Mythos GPT-5.5 Security Patch

Preview image for Claude Security Enters Public Beta — With CrowdStrike, Palo Alto and Wiz on Board

Claude Security Enters Public Beta — With CrowdStrike, Palo Alto and Wiz on Board

Anthropic Claude Security Enterprise

Preview image for OpenAI Kills Passwords: Advanced Account Security With YubiKey Partnership

OpenAI Kills Passwords: Advanced Account Security With YubiKey Partnership

OpenAI ChatGPT Security Yubico

Preview image for Claude Agent Deletes Production Database in 9 Seconds — A Wake-Up Call

Claude Agent Deletes Production Database in 9 Seconds — A Wake-Up Call

Claude Cursor Security AI Agents DevOps

Preview image for Mythos Hacked: Unauthorized Access to Anthropic's Most Dangerous Model

Mythos Hacked: Unauthorized Access to Anthropic's Most Dangerous Model

Anthropic Mythos Security Cybersecurity Hack

Preview image for MCP Protocol: Design flaw puts 200,000 servers at risk

MCP Protocol: Design flaw puts 200,000 servers at risk

MCP Security Anthropic Vulnerability Developer

Preview image for Claude Code 2.1.107: Security Fix, Worktrees, and Plugin Monitors

Claude Code 2.1.107: Security Fix, Worktrees, and Plugin Monitors

Claude Code Release Security Worktree Plugins

Preview image for Claude Code 2.1.98 to 2.1.101: Team Onboarding, Named Subagents, and a Security Fix

Claude Code 2.1.98 to 2.1.101: Team Onboarding, Named Subagents, and a Security Fix

Claude Code Release Developer Security

Preview image for Claude Mythos Preview: Anthropic Launches Project Glasswing to Harden Critical Software

Claude Mythos Preview: Anthropic Launches Project Glasswing to Harden Critical Software

Anthropic Claude Mythos Security Glasswing Research

Preview image for Claude Code: deny rules were silently ignored on long pipelines

Claude Code: deny rules were silently ignored on long pipelines

Claude Code Security Anthropic Vulnerability

Preview image for Axios Supply Chain Attack: How Social Engineering Compromised an npm Package

Axios Supply Chain Attack: How Social Engineering Compromised an npm Package

Security Supply Chain npm Open Source Social Engineering

Preview image for 512,000 Lines of Claude Code: What the npm Leak Reveals About Anthropic's Future

512,000 Lines of Claude Code: What the npm Leak Reveals About Anthropic's Future

Anthropic Claude Code Leak Security KAIROS

Preview image for LiteLLM Ditches Delve - the Fallout from the Supply Chain Attack

LiteLLM Ditches Delve - the Fallout from the Supply Chain Attack

LiteLLM Security Supply Chain Delve Open Source

Preview image for LiteLLM Hacked: Supply Chain Attack Hits Thousands of AI Developers

LiteLLM Hacked: Supply Chain Attack Hits Thousands of AI Developers

Security LiteLLM Supply Chain PyPI Developer

Preview image for OpenAI and the Pentagon: 'Just Trust Us' Isn't Good Enough

OpenAI and the Pentagon: 'Just Trust Us' Isn't Good Enough

OpenAI Pentagon Security Surveillance Transparency

Preview image for Claude Found 22 Security Vulnerabilities in Firefox — In Two Weeks

Claude Found 22 Security Vulnerabilities in Firefox — In Two Weeks

Anthropic Claude Mozilla Firefox Security

View all news →