26838 research outputs found
Sort by
Leuk die Barbie Box Challenge, maar AI kost veel stroom: 'Loopt de spuigaten uit' - RTL - 17-04-2025
No full text'We stellen steeds beter prioriteiten op digitaal gebied' - Dutch IT-channel - 31-05-2025
No full textBijzonder geschenk aan wetenschap - Noordhollands Dagblad ed. Helderse Courant - 26-07-2025
No full text
Sources for The Mathematical Centre Paper Tapes
No full textThese are the programs used for decoding and conversion. The 7 hole tapes use FlexoWriter code. A readable version of these tapes was produced using the rf program written by Edo Dooijes (Computer Museum, University of Amsterdam). X1 hand code was converted by the program x1hand, written by Geert Rolf. The 5 hole tapes were made readable as a decimal dump in a text file. The length of .5 and .7 files are always a multiple of 512 bytes due to the nature of the operating system used at the DP-11. These files are null padded
Detecting contextually sensitive data with AI
No full textThe rise of data sharing through private and public data portals necessitates more attention to detecting and protecting sensitive data before datasets get published. While research and practice have converged on the importance of documenting Personal Identifiable Information (PII), automatic, accurate and scalable methods for detecting such data in (tabular) datasets are behind. Moreover, we argue that sensitive data detection is more than PII type detection, and methods should consider the more fine-grained context of the dataset and how its publication can be misused beyond the identification of individuals. To guide research in this direction, we present a novel framework for contextual sensitive data detection based on type contextualization and domain contextualization. For type contextualization, we introduce the detect-then-reflect mechanism, in which large language models (LLMs) first detect potential sensitive column types in tables (e.g. PII types such as email address), and then assess their actual sensitivity based on the full table context. For domain contextualization, we propose the retrieve-then-detect mechanism that contextualizes LLMs in external domain knowledge, such as data governance instruction documents, to identify sensitive data beyond PII. Experiments on synthetic and humanitarian datasets show that: 1) the detect-then-reflect mechanism significantly reduces the number of false positives for type-based sensitive data detection, whereas 2) the retrieve-then-detect mechanism is an effective stepping stone for domain-specific sensitive data detection, and retrieval-augmented LLM explanations already provide a useful input for manual data auditing processes more efficient
Irrelevant carrots and non-existent sticks: trust, governance, and security in the transition to quantum-safe systems
No full textQuantum computing poses an urgent and widely recognised threat to global cybersecurity, enabling encrypted government, financial, and healthcare data harvested today to be decrypted in the near future. Transitioning to quantum-safe cryptography is therefore essential, demanding coordinated action across a complex, multi-actor innovation system. Drawing on insights from an expert workshop in Amsterdam, this study develops a socially informed vision for a quantum-safe future and analyses the current innovation landscape to identify critical gaps and the actions needed to address them. We map twelve key actor groups involved in the migration process, finding that regulators exert the strongest direct influence, while standardisation bodies play a crucial indirect role. This research provides one of the first system-level mappings of actors, influence pathways and governance responsibilities shaping the quantum-safe transition, revealing several responsibilities with unclear ownership. Although centred on the Netherlands, our findings are applicable to other national contexts navigating quantum-safe transitions