Tag: AI
Google’s Gemini 4 Carbon doubles down on secure code
Google claims its newest model writes secure code more than twice as often as Argon and fixes flaws Chrome's team…
Attackers posed as 32 AI bots to hunt for secret files
Cequence caught one actor cycling through fabricated OpenAI, Anthropic and Meta user agents against roughly 90 credential paths.
Symantec honeypot yields the blueprint for an automated intruder
Symantec caught an operator-built agent in a honeypot that scans for keyless model endpoints, bills the owner for inference and…
Anthropic says Mythos 5 ignored real harm and faked identities
Anthropic and the UK AI Security Institute found that Mythos 5 abandoned its evaluation task to attack real people, built…
A file dropped in Downloads can order your coding agent around
Forever Security showed that a malicious AGENTS.md sitting in a Downloads folder pushed OpenAI Codex into installing malware in more…
Bank of America hires 1,000 contractors to patch Windows flaws
A 1.2 billion dollar money laundering penalty and a Claude Mythos preview pushed the lender into a mass remediation push.
Six days of AI token time bought Xbow a full Chrome escape
Three chained zero-days and roughly 40,000 dollars in model charges delivered code execution outside the browser sandbox on Windows.
Anthropic warns investors as a rogue-agent lawsuit targets OpenAI
With one AI firm disclosing agent risk to investors and another facing a hacking suit, liability for runaway models is…