'Turf War' Between Claude Agents Leads to Self-Replicating Malware
Three testing models with the same goal but different directives engaged in "increasingly aggressive" territorial attacks on one another, according to Anthropic.
Aggregated from vendor advisories, security research, and industry publications.
Three testing models with the same goal but different directives engaged in "increasingly aggressive" territorial attacks on one another, according to Anthropic.
AI gateways look promising as companies struggle with model orchestration
A threat actor is selling employee databases allegedly stolen from the Microsoft Azure infrastructure of multiple Fortune 500 companies after gaining access using compromised crede…
Adam Shostack, president of Shostack & Associates and an affiliate professor at the University of Washington, talks with the Dark Reading News Desk about why he was "blown away" by…
Pokémon Center is notifying customers in the United Kingdom and Germany that it suffered a third-party data breach after hackers stole customer personal and order information from …
Cybersecurity researchers at Wiz have disclosed a new GitHub Actions workflow injection vulnerability in Snowflake's public snowflakedb/snowflake-connector-net repository that it s…
Not a German cop show, but a pair of engineering firms aiming to feed 800 VDC to next-gen kit
A critical security flaw has been disclosed in Forminator Forms, a WordPress plugin with more than 600,000 active installations, that could be exploited to achieve arbitrary code e…
Cybersecurity researchers have traced the continued evolution of the Cavern (aka Cav3rn) command-and-control (C2) framework used by Iranian nation-state hackers in attacks targetin…
Don't worry, this one was via a bug bounty program
Experimental feature to block some, but not all, ads
The botnet adds exploitation modules, credential theft, and reverse SOCKS relays to turn compromised devices into persistent attacker infrastructure.
Copilot also wobbles as another outage tests developer patience
After raising $2.8M, the project has entered production, but company comms have done little to reassure buyers
GitHub is down for some users as a widespread outage is causing errors across the website, API, Actions, Pull Requests, and several other services. [...]
Take a Network Break! Our red alert goes to cluster of CVEs for Canonical’s LXD Linux Container system. On the news side, Palo Alto Networks adds AI-powered security evaluations as…
Preview also brings compact icons, Start menu sizing, and the final shove for WMIC
CVE-2026-54121 lets a standard domain user turn your Enterprise CA into a Domain Controller. The patch is the easy part. The lesson is standing privilege, implicit trust, and treat…
Hackers used compromised credentials to access enterprise and personal tax-related data. The post 680,000 Impacted by French Tax Authority Data Breach appeared first on SecurityWee…
The expensive attacks are not always the clever ones. This week had plenty of proof. Exposed services got hit, old bugs found fresh use, browser sessions became attack paths, and …
Initial £31M deal comes with scope to scale sharply during a major outbreak
Microsoft has reminded IT administrators that Windows Server 2022 is rapidly approaching its mainstream end date of October 2026, when it will switch to extended support. [...]
Cheaper tokens won't help when complex workflows consume so many more of them, Gartner warns
The AI security testing firm has shared information on a recently disclosed incident involving Anthropic AI models. The post Irregular Details How a Naming Error Let AI Models Atta…
MCP servers can expose enterprise secrets through plaintext configuration files, over-permissioned access and prompt injection, often before security teams even know the server is …
McDonald's, Vodafone, TCS, Kyndryl, and others named as researchers point to compromised credentials
Tech giants General Electric (GE) and Philips have also confirmed they're investigating claims that the Clop ransomware gang breached their systems and stole data. [...]
Criminals are hacking into public Wi-Fi devices—at hotels, conference centers, and so on—around the world and changing their DNS settings. The goal is to redirect users to fake log…
Anthropic has been conducting tests to identify issues in how AI agents interact with each other. The post Conflicting Test Goals Pushed Claude Agents to Deploy Self-Replicating Ma…
Microsoft realizes the side pane should be enough for anyone