The Hidden Instructions That Can Hijack AI Agents
Malicious prompts concealed in documents, metadata, emails, images and code can manipulate autonomous agents into taking
#news
共收录 3762 条相关情报。
← 返回所有主题Hackers Return $263 Million Stolen From Liquid Network
Alleged ‘white-hat’ hackers drained $320 million from Liquid’s federation wallet, demanding a bug fix. The post Hackers
ShinyHunters hackers claim breach of Florida "DAVID" DMV database
The ShinyHunters extortion gang claims it breached an online platform for the Florida Department of Motor Vehicles datab
OpenAI says ChatGPT outage causes image generation errors
OpenAI is investigating an ongoing incident causing ChatGPT image generation failures and delays when uploading files. [
Slim Spider Steals Crypto Custody Secrets From Brazilian Financial Institution
A previously undocumented financially motivated threat actor has been linked to attacks targeting Brazilian financial in
August updates trigger 0xc0000409 errors on Windows Server 2016
Microsoft says the August 2026 security update may trigger 0xc0000409 errors on Windows Server 2016 systems where the Co
Cylake Raises $245 Million Ahead of Cybersecurity Platform Beta
The startup founded by Palo Alto Networks’ Nir Zuk has raised $290 million to build an AI-native security platform for h
SAP warns of maximum severity 'OVERPASS' kernel vulnerability
SAP has addressed 20 vulnerabilities across multiple products in its September 2026 security updates, including a maximu
SAP Patches Critical Extended Passport Processing Vulnerability
Affecting the SAP kernel code, the flaw allows unauthenticated, remote attackers to run arbitrary commands, recover secr
Liquid Hackers Return 3,400 Bitcoin Taken via Elements Bug, Still Holding $47M in BTC
Whoever took nearly 4,000 bitcoin from the Liquid Network on Sunday, September 6, returned 3,400 of it the next day, Bit
OpenAI says GPT-6 Astra can find zero-days, but is also harder to monitor
OpenAI confirmed that GPT-6 Astra is the first model it has broadly deployed to reach the "Critical level" for cybersecu
ChatGPT Flaw Let a Planted Prompt Send a Victim's Gmail Data to Another Account
Check Point Research said in a report published today that a single instruction planted in a ChatGPT conversation could
Autonomous AI Agents Compromise Thousands of Credentials in Under Six Hours
Threat actors are continuing to leverage artificial intelligence (AI) to streamline their operations, with one financial
Adobe fixes critical Magento zero-day exploited to backdoor servers
CVE-2026-75650
Adobe has released an emergency fix for CVE-2026-75650, an actively exploited max-severity zero-day vulnerability dubbed
Party’s Over for Crypto Scammers Who Went on a Spending Spree After a $240 Million Bitcoin Theft
The scammers purchased fleets of sports cars, flew on private jets, hired security guards and rented mansions in Miami a
Webinar: The forgotten Google Workspace access that can lead to a breach
Third-party applications connected to Google Workspace can retain access long after their original purpose is forgotten.
Hackers build AI frameworks for widescale credential theft
Threat actors are increasingly switching from AI-powered coding assistants to multi-agent frameworks that automate every
WeChat Zero-Click Worm Took Over Accounts on iPhone and Android via Incoming Calls
Researchers at the security firm Calif have built a worm that takes over a WeChat account via an incoming call and demon
What It Took to Reach 1 Billion Build Manifests
In the last six months, Chainguard doubled its output from 500 million to more than 1 billion container build manifests.
FreeIPA Flaw Chain Lets Anonymous Clients Create Reusable Administrator Credentials
A flaw in FreeIPA lets a client that has never logged in create a Kerberos identity of its own choosing in the directory