Robuta

Prompt Injections gehören zu den größten Sicherheitsrisiken von KI. Sie schleusen unbemerkt Manipulationen ein und können Daten verfälschen oder...

Anthropic is an AI safety and research company that's working to build reliable, interpretable, and steerable AI systems.

Nov 25, 2025 - Claude Opus 4.5 scores higher than its rivals in prompt-injection security, but the results show how limited these defenses still are.

Nov 25, 2025 - Claude Opus 4.5 ist laut Hersteller Anthropic das am besten gesicherte Modell gegen Prompt Injections. Aber das heißt nicht viel.

Aug 25, 2025 - See how to reduce the risks of an indirect prompt injection, such as the exposure of confidential files or the execution of code without the user's consent.

Learn about visual prompt injections, their appearance, and top defense strategies against these attacks.

Independent AI researcher Johann Rehberger (previously) has had an absurdly busy August. Under the heading The Month of AI Bugs he has been publishing one...

So, I saw this tweet. It turns out you can ask GPT4 to compress a message in a way it can understand. The end result is usually not understandable fo…

AI browsers remain vulnerable to prompt injection attacks via screenshots and hidden content, allowing attackers to exploit users' authenticated sessions.

This new paper by 11 authors from organizations including IBM, Invariant Labs, ETH Zurich, Google and Microsoft is an excellent addition to the literature on...

Claude API Documentation

Dec 23, 2025 - KI-Agenten und Browser werden besser gegen Prompt Injections geschützt. Doch: Das Problem wird laut OpenAI noch Jahre bestehen.

Jul 31, 2025 - Summary AI tools like Cursor are changing how software gets written, making coding faster, easier, and smarter. But HiddenLayer’s latest research reveals a...