AI safety vs AI ethics: what’s the difference?
How AI safety and AI ethics differ, where they overlap, and which questions each one answers before an AI system is put to use.
Research & Insights
Research, analysis and commentary on responsible AI, technology governance and human-centered innovation.
Publications
The Foundation’s technical reports, published in English.
- Bias Detection Under Article 4a of the EU AI Act: When Synthetic, Anonymised or Differentially Private Data Is Enough. Technical evidence options, a simulation study and a real-data replication
- Using Generative AI to Create Synthetic Medical Data for Patient Privacy Protection: A Review and an Empirical Evaluation of the Fidelity-Utility-Privacy Trade-offData package: Synthetic medical data and patient privacy benchmark (WHAS500, WDBC, FLCHAIN)
AI Safety
See all →What is AI alignment and why does it matter?
Learn what AI alignment means, why stated goals and real intentions drift apart, and how researchers try to close the gap before autonomous systems widen it.
What is AI safety?
A plain-language introduction to AI safety: what alignment, robustness, interpretability and control mean, and why human oversight needs more than a checkbox.
Responsible AI
See all →Building trust: transparency in AI decision-making
What meaningful transparency in AI decisions requires: disclosure, usable explanations, auditability and a real route to contest an outcome.
What is responsible AI?
What responsible AI means in practice: how fairness, transparency and accountability become concrete decisions before, during and after an AI system is used.
Technology Policy
See all →The future of AI regulation in Europe
What comes after the EU AI Act and its 2026 amendment: standards, codes of practice, enforcement, court rulings, and what builders and citizens can expect.
The EU AI Act: what it means for the next decade
How the EU AI Act sorts AI by risk, what it requires of providers and deployers, when each rule applies, and why its reach extends beyond Europe.
AI governance explained
How laws, standards, institutions and norms decide how AI may be built and used, and what that means for a school, a hospital or a city office.
Digital Society
See all →Human-centered AI: principles and challenges
What human-centered AI means in practice, why human oversight is harder than it sounds, and the trade-offs between automation, control and autonomy.
Bridging the AI divide: access as a human right
What the AI divide is, why access to AI is a rights question, and which design and policy choices decide whether AI narrows or widens social gaps.
Foundation Updates
Foundation Updates
First technical reports published
Two studies are available on ResearchGate: bias detection under Article 4a of the EU AI Act and synthetic medical data for patient privacy. More are in preparation.
Foundation Updates
First articles published
Our first articles on AI safety, responsible AI, technology policy and digital society are now available, each in six languages.
Foundation Updates
Looking for research partners
We are looking for researchers, institutions and partners to work with on safe, beneficial AI.
Foundation Updates
LASTHUMAN Foundation is officially registered
Established as a registered foundation (KRS 0001228457), with a clear mission for human-centered, responsible AI.