Fields
AI security
Prompt injection and jailbreaks, adversarial attacks, data poisoning and model theft, privacy attacks, agent and supply-chain security, AI for cybersecurity
Most mentioned companies
Stories: 4 · newest first
October 9, 2026
-
Anthropic launches long-term 'Cyber Mission' for defenders, offers free 'OSS Scanner' to open-source projects (report)
Anthropic announced 'Anthropic Cyber Mission', a long-term program providing tools, research, and resources to defenders who protect...
-
Cambridge launches Centre for Cryptographic Trust Infrastructure to let mutually untrusting AI agents verify facts (report)
The University of Cambridge's Department of Computer Science and Technology announced the Centre for Cryptographic Trust Infrastructure...
October 8, 2026
-
Red Hat and IBM used 'Lightwell' to fix 400+ undisclosed Java library vulnerabilities and launch 'Lightwell Clearinghouse' (report)
Red Hat and IBM used Lightwell to find and fix more than 400 previously unknown vulnerabilities in Java libraries. They also made...
-
Meta deploys AI tool to detect disguised ads that steer users to external child sexual abuse material (report)
Meta found ads on its platforms that looked ordinary but steered users to child sexual abuse material on external sites. In response, it...