惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

IntelliJ IDEA : IntelliJ IDEA – the Leading IDE for Professional Development in Java and Kotlin | The JetBrains Blog
IntelliJ IDEA : IntelliJ IDEA – the Leading IDE for Professional Development in Java and Kotlin | The JetBrains Blog
G
GRAHAM CLULEY
P
Privacy & Cybersecurity Law Blog
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
宝玉的分享
宝玉的分享
P
Proofpoint News Feed
H
Help Net Security
V
Visual Studio Blog
阮一峰的网络日志
阮一峰的网络日志
C
Cisco Blogs
人人都是产品经理
人人都是产品经理
Know Your Adversary
Know Your Adversary
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
Recorded Future
Recorded Future
I
Intezer
罗磊的独立博客
T
The Exploit Database - CXSecurity.com
Blog — PlanetScale
Blog — PlanetScale
Malwarebytes
Malwarebytes
Spread Privacy
Spread Privacy
T
Tor Project blog
V
Vulnerabilities – Threatpost
云风的 BLOG
云风的 BLOG
腾讯CDC
B
Blog RSS Feed
Stack Overflow Blog
Stack Overflow Blog
F
Future of Privacy Forum
MyScale Blog
MyScale Blog
Latest news
Latest news
IT之家
IT之家
MongoDB | Blog
MongoDB | Blog
The Hacker News
The Hacker News
S
Securelist
博客园 - 【当耐特】
C
CXSECURITY Database RSS Feed - CXSecurity.com
T
Threat Research - Cisco Blogs
Jina AI
Jina AI
Cisco Talos Blog
Cisco Talos Blog
B
Blog
博客园 - 三生石上(FineUI控件)
Last Week in AI
Last Week in AI
CTFtime.org: upcoming CTF events
CTFtime.org: upcoming CTF events
M
MIT News - Artificial intelligence
V
V2EX
D
Darknet – Hacking Tools, Hacker News & Cyber Security
The Cloudflare Blog
The GitHub Blog
The GitHub Blog
博客园 - 聂微东
F
Full Disclosure
C
CERT Recently Published Vulnerability Notes

Hackread – Cybersecurity News, Data Breaches, AI and More

Europol Seizes First VPN Used by Ransomware Gangs, Arrests Administrator Android Malware Spotted Subscribing Victims to Paid Services Without Consent Microsoft’s Retired IE Tool MSHTA Now Being Used in Fileless Malware Attacks Understanding Trend Structure: Higher Highs and Lower Lows Explained GitHub Breach: TeamPCP Steals 3,800 Repositories via VS Code Extension Verizon DBIR: AI Helped Hackers Exploit Vulnerabilities in 31% of Recent Breaches Fake Word Phishing Reveals Enterprise Blind Spot in Trusted Remote Access Tools Banana RAT Malware in Fake Invoices Hits Customers at 16 Brazilian Banks AI Agent Security: Automating Workflow Without Creating Prompt Injection or Data Leak Risks How Parts Inventory Management Software Fixes Inventory Challenges Pwn2Own Berlin 2026 Closes With $1.3 Million in Zero-Day Payouts Criminal IP Returns to Infosecurity Europe 2026 with Advanced AI-Driven TI & ASM Two-Thirds of Nonhuman Accounts Are Unseen and Unmanaged, According to Orchid Security's Identity Gap Hosting Service Standards That Define High-Performing Agencies Hackers Actively Exploit ‘Nginx Rift’ Vulnerability Affecting NGINX, F5 Products 10 Top OSINT Tools Every Investigator Should Know in 2026 New Reaper Malware Uses Fake Microsoft Domain to Steal macOS Passwords 10 Tips for Phrasing Employee Feedback in Reviews Government Backed Hackers Abuse Cloudflare in Malaysian Espionage Campaign Continuous Detection, Continuous Response: Mate Security Redefines the Modern SOC The Gentlemen Ransomware Gang Hit by Internal Breach, Operations Exposed Closing the Gap: The Regulatory and Structural Maturation of Digital Assets Scammers Send Physical Phishing Letters to Steal Ledger Wallet Seed Phrases Grafana Says It Rejected Ransom Demand After Source Code Theft AI Voice Cloning: The Technology Behind It, Who's Building It, and Where It's Headed Critical ‘Claw Chain’ Vulnerabilities Put Thousands of OpenClaw AI Servers at Risk The Next Cybersecurity Challenge May Be Verifying AI Agents Hackers Use PyInstaller and AMSI Patching to Deliver XWorm RAT v7.4 CalPhishing Scam Uses EvilTokens Kit, Outlook Invites to Steal M365 Sessions Fake Job Interview Apps Drop JobStealer Malware on Windows and macOS How Fintech APIs Are Modernizing Business Cash Flow Management FamousSparrow Targeted Oil and Gas Industry via MS Exchange Server Exploit China-Linked Twill Typhoon Uses Fake Apple and Yahoo Sites for Espionage TeamPCP Claims Sale of Mistral AI Repositories Amid Mini Shai-Hulud Attack Instructure Reaches Deal with ShinyHunters to Prevent Canvas Data Leak TeamPCP Used Mini Shai-Hulud Worm to Poison Over 400 npm and PyPI Packages Slovakian Admin of Dark Web Kingdom Market Jailed for 16 Years in US Why Canadian Telecom Providers Are Prime Targets for Cyberattacks Canvas Hackers ShinyHunters Say Their Official Domain Was Suspended Fake Claude Code Installer Targets Developers With Browser Credential Stealer Pwn2Own Berlin 2026 Hits Capacity as Rejected Hackers Release 0-Days Top Video Downloaders in 2026: Why Wondershare UniConverter Remains a Strong Choice Operation HumanitarianBait Uses Fake Aid Documents to Deploy Python Spyware Google Says Hackers Used AI to Develop a Zero-Day Exploit Romanian Man Faces Up to 30 Years in US Prison Over Vishing Scams 9-Year-Old Dirty Frag Vulnerability Enables Root Access on Linux Systems Lyrie.ai Joins First Batch of Anthropic’s Cyber Verification Program Hackers Exploit Vercel GenAI to Mass-Produce Convincing Phishing Sites Two US Men Sentenced for Helping North Korean Hackers Infiltrate US Firms Hackers Trick DigiCert Into Issuing Certificates Used to Sign Malware Hackers Hijack JDownloader Site to Deliver Malware Through Installers Fake macOS Troubleshooting Sites Used to Steal iCloud Data in ClickFix Scam ClaudeBleed Vulnerability Lets Hackers Hijack Claude Chrome Extension to Steal Data ShinyHunters Defaces Canvas LMS Portal, Hundreds of Universities Affected Hackers Use Fake Claude AI Site to Infect Users With New Beagle Malware Researcher Shows Edge Browser Stores Saved Passwords in Plaintext Google Chrome Accused of Silently Installing 4GB AI Model on User Devices Why Outdated Maintenance Software Is a Growing Ransomware Risk Scammers Use Hidden Text to Bypass AI Email Filters in Phishing Scams Best OSINT Tools for Investigations and Threat Intelligence in 2026 Google Fixes CVSS 10 Gemini CLI Vulnerability Enabling GitHub Issue-Based RCE ShinyHunters’ Instructure Canvas LMS and Vimeo Breaches Impact Millions of Users Building Strategic Advantage With Integrated Planning The "Juice" Factor: Designing Game Feel Application Security Strategies Are Changing as AI-generated Code Floods the SDLC Massive “Low and Slow” DDoS Attack Hits Platform With 2.45 Billion in 5 Hours LuxSci Launches Enterprise-Grade HIPAA-Compliant Email Security for Mid-Sized Healthcare Organizations Anti-ICE Site GTFO ICE Accused of Exposing Data of 17,000+ Activists FEMITBOT Network Abuses Telegram Mini Apps for Crypto Scams and Android Malware Wiz ZeroDay.Cloud Event Reveals 20-Year-Old PostgreSQL Vulnerabilities Cyber-Secure Philanthropy: Tech Infrastructure for Global Donations 7 Key Features That Make Secure Browsers Safer Paying Ransom Won’t Help as VECT 2.0 Ransomware Destroys Data Irreversibly Google AppSheet Exploited in 30,000-User Facebook Phishing Operation 2 US Cybersecurity Experts Jailed for Aiding ALPHV (BlackCat) Ransomware 45,000 Attacks, 5,300+ Backdoors Tied to China-Linked Cybercrime Operation Hackers Use Jenkins Access to Deploy DDoS Botnet Against Gaming Servers Criminal IP and Securonix ThreatQ Collaborate to Enhance Threat Intelligence Operations Critical cPanel Vulnerability Lets Attackers Bypass Login, Gain Root Access Best Diagram Software in 2026, Why EdrawMax Works for Everyday Use Private Chats, Photos of Celebs Exposed in Suspected Stalkerware Leak Misconfigured Server Run by Hackers Leaks 345,000 Stolen Credit Cards Managed vs Self-Managed Cloud Hosting: Choosing the Best Option for Your Business 9-Year-Old Linux Kernel Vulnerability “Copy Fail” Enables Full Root Access Cursor AI Agent Wipes PocketOS Database and Backups in 9 Seconds New AI-Powered Bluekit Phishing Kit Targets Major Platforms with MFA Bypass Attacks Polymarket Rejects Data Breach Claims as Hacker Alleges 300K Records Stolen Brinker Introduces a Novel Approach to Deepfake Detection US-Estonian Suspect Arrested Over Alleged Scattered Spider Cyberattacks Cursor AI IDE Vulnerability Allows Code Execution Via Hidden Git Hooks Top AI-Powered Vendor Risk Management Platforms for SaaS Companies in 2026 New DHL Phishing Scam Uses 11-Step Attack Chain to Steal Passwords Decoding Q1 2026's $152.9 Billion Crypto Custody ConcentrationDecoding Q1 2026's $152.9 Billion Pack2TheRoot: 12-Year-Old Linux PackageKit Flaw Enables Full Compromise Stablecoins: Always-On Money Needs Always-On Controls New Linux FIRESTARTER Backdoor Targets Cisco Firepower Devices Why Unofficial Download Sources Are Still a Security Risk in 2026 The Role of Aggregated Liquidity in Modern Crypto Markets 82 Chrome Extensions Found Selling User Data, 6.5 Million Users Affected ShinyHunters Leaks Data of Udemy, Zara, 7-Eleven in Salesforce Linked Breach
Deleted Google API Keys Remain Active up to 23 Minutes, Study Finds
Deeba Ahmed · 2026-05-22 · via Hackread – Cybersecurity News, Data Breaches, AI and More

A new study conducted by the cybersecurity firm Aikido Security reveals that deleted Google API keys stay active and can continue authenticating successfully for up to 23 minutes after they are removed. The results were obtained after running 10 controlled trials over two days to measure the delay.

Key Findings

An API key is a string of data used to authenticate requests between software applications. According to researchers, the Google Cloud Platform (GCP) console shows the key as deleted immediately. However, tests showed that the keys actually take an average of 16 minutes to stop working completely, with the longest delay lasting nearly 23 minutes.

During this timeframe, threat actors holding a leaked key retain full access to any enabled APIs on the project. This allows them to exfiltrate cached conversations and dump files uploaded to Gemini. They can also access BigQuery data and Maps APIs.

Why Does The Issue Occur?

In the blog post, which was published today and shared exclusively with Hackread.com, researchers explained that this issue happens because of eventual consistency in Google’s authentication infrastructure. In this distributed systems model, updates propagate gradually across global servers rather than all at once.

This means when you delete a key, the message does not reach every Google server around the world immediately, giving hackers a temporary gap to use the key on servers that haven’t updated yet. This identical class of infrastructure issue was demonstrated on AWS last year by researcher Eduard Agavriloae, though the AWS revocation window was only 4 seconds.

Tracking and Infrastructure Differences

The attack method relies on the hacker sending continuous authenticated requests to rotate through Google’s global authentication servers before they sync. Testing across different GCP regions revealed diverse regional variations. In the first minute after deletion, virtual machines in the asia-southeast1 region saw a median success rate of 22%, while us-east1 and europe-west1 both allowed 49% of requests to succeed.

For incident response teams, tracking the timeline of events during an attack is complicated by the GCP “Traffic by Credential” graph. When a key gets deleted, any further authentication attempts by an attacker are bundled into a generic category labelled apikey:UNKNOWN. This makes it difficult to pinpoint which specific credential an attacker is trying to misuse.

Watch the demo

Researchers noted that Google has already solved faster propagation for other credential types. For example, Google Service Account keys revoke in roughly 5 seconds, while newer Gemini-format keys (which use an AQ. prefix) take about 1 minute.

Aikido Security reported these findings to Google, but the company closed the report as “won’t fix,” stating that propagation delay is a known property of the system and not a security flaw. Consequently, researchers advise treating Google API key deletion as a 30-minute operation and monitoring the GCP console for valid authentications within that window.