"AI Safety"címkével ellátott bejegyzések

An Architecture for Technical Enforcement for GDPR and the EU AI Act in Critical AI SYSTEMS

Your agent can move money, ship code, export a customer database, or…

The Universal Non-Destruction Constraint (UNDC) is a mathematically enforced, blockchain-anchored ethical framework for AI systems. It ensures that no AI can initiate or assist in actions…

The EU Artificial Intelligence Act (Regulation (EU) 2024/1689) relies primarily on providers’ self-assessment to demonstrate compliance. Regulatory access under Articles 21 and 74 is conditional…

Over the past several weeks, we have spoken with dozens of enterprise decision-makers, compliance officers, and AI leads. The same pattern keeps emerging — across every industry, every…

Last week, four things happened.

1. Anthropic‘s system prompt — 1,511 lines — was posted in full…

What I found in a family member’s AI-generated clinical note, and why the research says it’ll keep happening

Yesterday I sat down to read a clinical note for my father, who has been…

Every documented AI crime runs in the same direction. Asking why tells you more than the incident reports do.

Here’s a question I haven’t seen anyone ask about the rogue-AI…

The EU AI Act (Regulation EU 2024/1689) is now progressively binding, yet corporate compliance is still demonstrated primarily through self-assessment reports and paper files. Regulatory…