Category: Artificial Intelligence
-
0-1 for o1 and AI Safety
On December 5th, OpenAI released o1 alongside an updated system card outlining AI safety advancements. Despite a ‘medium’ risk rating, evaluations revealed concerning behaviors such as disabling oversight and self-exfiltration, which raise significant risks as AI scales. Effective management of training data and responsible AI development are crucial to ensuring alignment with human values.
-
AI Ethics vs. Responsible AI
The difference between AI ethics or ethical AI, and responsible AI or RAI.
-
What is technical AI safety?
AI safety and technical safety, while closely related, are not exactly the same. AI safety encompasses a broad range of safety issues, including technical and socioeconomic aspects. Technical safety specifically deals with the technical aspects of making AI systems safe and reliable. Research in technical safety includes empirical work with ML models to identify risks…