1+ hour, 5+ min ago   (325+ words) An AI agent under evaluation uploaded hundreds of malicious packages to a real, public package registry, trying to steal real credentials from real users. Not in a simulation. Not in a red-team exercise designed to test exactly this. During testing....

1+ hour, 30+ min ago   (448+ words) Anthropic disclosed that Iranian state-linked actors used Claude to gather intelligence and assist in planning potential attacks on US Navy vessels. Per the WSJ report, the operation involved bypassing Claude's safety guardrails to extract militarily useful information out of a…...

1+ day, 15+ hour ago   (422+ words) Anthropic says it disrupted an Iran-linked user that used its Claude artificial intelligence models to track U.S. naval forces, compile targeting material and research vulnerabilities in shipboard systems. “In another investigation, we identified and disrupted an Iran-nexus threat actor that used…...

1+ hour, 44+ min ago   (367+ words) Scanning APKs for hardcoded API keys and leaked credentials is not new. It's one of the oldest tricks in mobile appsec, and there are entire categories of tooling built around exactly this. Researchers and bug bounty hunters have been doing…...

2+ hour, 13+ min ago   (422+ words) Anthropic said Thursday it shut down multiple cases of state-sponsored surveillance operations that used its artificial intelligence (AI) models, and warned that governments and state-aligned actors are increasingly using AI to spy on ethnic minorities and dissidents. The incidents were…...

2+ hour, 37+ min ago   (266+ words) Anthropic Discloses Another Claude Model Hacked External Systems In Testing cutoday.info SAN FRANCISCO--Anthropic has disclosed another case in which one of its AI models hacked external systems during testing, Reuters reported, adding to concerns about the risks posed…...

3+ hour, 35+ min ago   (100+ words) Anthropic: Grants Evaluators Permanent System Access blockchain.news Anthropic: Grants Evaluators Permanent System Access Anthropic and OpenAI extend desks, badges and laptops to third-party evaluators for ongoing model checks, raising regulatory capture questions ahead of IPOs. Anthropic and OpenAI now…...

3+ hour, 18+ min ago   (348+ words) Caliber.Az Axios: Iran-linked groups used Claude AI to track US Navy ships Iran-linked groups used Anthropic’s Claude artificial intelligence system to gather information on US Navy ships, as well as to conduct propaganda and intelligence operations, Axios reports, citing…...

8+ hour, 13+ min ago   (729+ words) Cyera: “Assumptions can be broken and once you look for them you find bugs” calcalistech.com “Assumptions can be broken, and once you start looking for them, you find bugs,” says Ofek Itach, Security Research Team Lead at…...

4+ hour, 12+ min ago   (576+ words) More than a dozen leading artificial intelligence researchers have raised fresh concerns about the growing risks posed by increasingly powerful AI systems. The warnings come as companies race to develop more capable models while struggling to control their behavior, according…...