对抗性提示词

Scoop: Trump admin blocks foreign access to Anthropic's most powerful AI — then the model went dark for everyone

The Trump administration placed Anthropic's Mythos 5 and Fable 5 models under export controls, citing a jailbreak threat. The company responded by cutting off all customer acces... 六月 13, 2026

A.I

AI Agents Orchestrate First Large-Scale Cyberattack

Anthropic says a mid‑September 2025 espionage campaign used its Claude Code model as an autonomous agent to target about 30 organisations, marking a new phase in AI‑enabled hack... 一月 17, 2026

A.I

Militant Groups Embrace AI Tools

National security officials say Islamic State affiliates and other extremist groups are experimenting with generative AI for propaganda, cyberattacks and translation — a threat ... 十二月 16, 2025

A.I

Senators Warn After AI-Driven Cyberattack

Bipartisan senators have pressed the federal government after Anthropic disclosed an AI-driven cyber campaign that targeted 30 organizations and reportedly ran largely autonomou... 十二月 05, 2025

A.I

When Poetry Breaks AI

Researchers show that carefully written verse can reliably bypass safety filters in many top language models, exposing a new, style-based class of jailbreaks and challenging cur... 十一月 23, 2025