Modern IT systems are characterized by their ever-increasing complexity. In order for IT security to keep up with this, automation needs to be further developed and completely rethought. Artificial intelligence (AI) methods are instrumental in this process and can support humans in analyzing and protecting security-critical systems. However, just like conventional IT systems, AI systems can be attacked. The main challenge here is to find and fix any vulnerabilities in the algorithms.
How can machine learning respect privacy without sacrificing fairness? Discover DPPL, a prototype-based method that provides strong privacy guarantees while boosting accuracy for underrepresented groups. By addressing bias in differentially private models, this approach ensures ethical and inclusive AI development without compromising performance.
Deepfakes are a significant threat to democracy as well as private individuals and companies. They make it possible to spread disinformation, to steal intellectual property and to commit fraud, to name but a few. While robust AI detection systems offer a possible solution, their effectiveness depends largely on the quality of the underlying data, simply put: »Garbage in, garbage out.« But how do you create a dataset that is well suited to identifying the ever-evolving deepfakes and enables robust detection? And what constitutes high-quality training data?
ChatGPT is the AI software that supposedly does it all: It’s expected to compose newspaper articles and write theses — or program malware. Is ChatGPT developing into a new tool for hackers and cyber criminals that makes it even easier for them to create malware? Institute director Prof. Dr. Claudia Eckert and AI expert Dr. Nicolas Müller give their opinion on the potential threat to digital security posed by ChatGPT.
Machine learning is being hailed as the new savior. As the hype around artificial intelligence (AI) increases, trust is being placed in it to solve even the most complex of problems. Results from the lab back up these expectations. Detecting a Covid-19 infection using X-ray images or even speech, autonomous driving, automatic deepfake recognition — all of this is possible using AI under laboratory conditions. Yet when these models are applied in real life, the results are often less than adequate. Why is that? If machine learning is viable in the lab, why is it such a challenge to transfer it to real-life scenarios? And how can we build models that are more robust in the real world? This blog article scrutinizes scientific machine learning models and outlines possible ways of increasing the accuracy of AI in practice.
How secure is artificial intelligence (AI)? Does a machine perceive its environment in a different way to humans? Can an algorithm’s assessment be trusted? These are some of the questions we are exploring in the project “SuKI — Security for and with artificial intelligence”. The more AI is integrated into our everyday lives, the more important these questions become: When it comes to critical decisions — be it on the roads, in the financial sector or even in the medical sector — which are taken by autonomous systems, being able to trust AI is vital. As part of our ongoing SuKI project, we have now successfully deceived the state-of-the-art object recognition system YoloV3 [0].