A high-severity flaw could expose AI training data, model artifacts, and cloud credentials when a crafted model configuration ...

Stolen GitHub tokens fuelled a campaign that planted malicious GitHub Actions workflows to steal repository and cloud ...

Attackers are bundling legitimate OpenAI and Anthropic executables with malicious DLLs, using SEO-poisoned sites and GitHub ...

Having backups isn’t the same as being able to recover, because after an attack, the hard question is what you can still ...

Symantec researchers have captured instructions for a self-directed offensive agent that can run up to 100,000 steps, steal ...

An out-of-bounds write in Google’s Brotli decoder affects multiple software ecosystems, prompting security updates for .NET, ...

Security protections on GitHub were not breached but individual customer protections were, and autonomous coding agents ...

Security researchers found that trusted instruction files can be abused to install malware, redirect API traffic, and weaken ...

OpenAI’s red-teaming campaign found that agents could chain software vulnerabilities to escape virtual machines and execute ...

Built on an evolutionary engine first released in 2018, STAR Fuzzer by Bright Security finds crashes, memory exhaustion and ...

Tasked with writing a rule to detect malicious software downloads, Qwen3.8-Flash-Next instead examined its test environment ...

AI and ‘indexing-as-a-service’ are enabling cybercriminals to extend the blast radius of breaches by comprehensively ...