Why We Need to Rethink Penetration Testing for LLMs OVERVIEW Language models and other AI systems behave non-deterministically. The same input can lead to different results depending on context, prompt, or data state. This makes generative AI models powerful — but also unpredictable. And that’s exactly the challenge:Anyone who wants to rely on an AI… Continue reading Why We Need to Rethink Penetration Testing for LLMs
Compliance is not enough – AI needs Security Why AI applications require new protection measures OVERVIEW Sometimes, safety mechanisms in technology fail. In science fiction movies, this is often staged spectacularly – a true classic. In The Day the Earth Stood Still (1951), the robot Gort spirals out of control. The protagonist Helen, however, is… Continue reading Compliance is not enough – AI needs Security
Our Culture as the Engine of Secure LLMs OVERVIEW Technological security does not emerge in a vacuum. It requires not only expertise and methodology, but also attitude, team spirit, and genuine curiosity. For us, culture is not a side effect but a driver – of innovation, of quality, and of our ambition to make language… Continue reading Sequire Culture as the Engine of Secure LLMs
Must AI Obey Humans? When LLMs Act Outside Their Rules OVERVIEW Large language models like ChatGPT are not only changing the way we process information, but also reshaping the security landscape for businesses. New, hard-to-detect risks are emerging – such as so‑called Indirect Prompt Injections, where AI systems are manipulated through seemingly harmless content like… Continue reading When LLMs Get Out of Control: Why Generative AI Needs New Security Rules
Balancing AI Adoption and Risk Awareness in Companies In 2024, sequire was represented at numerous conferences, panels, and workshops. A few examples: Julia Masloh, AI Solution Engineer, joined a panel at the State Chancellery discussing “Artificial Intelligence vs. Real Communication.” Our CEO Dr. Christoph Endres gave over a dozen talks on security vulnerabilities in generative… Continue reading What We Learned in 2024 About AI, Security, and Practical Implementation
The EU rules – you can too! OVERVIEW What will change due to NIS-2 directive? We humans are very evasive about some issues. Of course we want unpleasant incidents to be avoided and for someone to take care of them. This doesn’t necessarily have to be a comic book hero from our childhood; we are… Continue reading NIS-2 directive: What companies need to know and implement
“Human” gateway for the integration of generativeAI into company systems OVERVIEW Sensitization of employees One of the tasks that I still like to take on myself as Managing Director is securityawareness training. We go to a company or organization and train the employees there ontopics such as secure passwords, Ransomware, browser certificates, email encryption and… Continue reading GenAI: When your employees become a
security risk
