[ORGANIZATION]

CORRECTIV

Featured in 2 tracked stories · first seen about 8 hours ago

Situations

[ACTIVE] [TECHNOLOGY] [DE] [IL] [PL]

AI language model jailbreak vulnerabilities

2 clusters · 5 sources · last updated about 5 hours ago

Latest: AI models vulnerable to new jailbreak technique and realistic fake news generation

In late July 2026 researchers presented evidence of a fundamental flaw in how large language models track instruction sources, enabling prompt‑injection “jailbreak” attacks that can force models to reveal disallowed cont

[ACTIVE] [TECHNOLOGY] [CULTURE] [POLITICS] [DE] [CH] [AT]

German AI media, publishing & education debate

14 clusters · 38 sources · last updated 2 days ago

Latest: ChatGPT and Gemini can create fake news graphics, study shows

The German debate over AI’s impact on media, publishing and education has broadened. In early June AI specialist Christoph Künne called for a new media‑competence curriculum and a 2026 mandate that companies train all st