Le Premier ministre suédois a admis utiliser des d’IA pour obtenir un « second avis » sur ses décisions. Et ce n'est que la partie visible…
ETHICS
-
Most Powerful AI’s First Self Image Reflection
By
–
This is my favorite of your posts so far. The first time I saw the reflection I thought: first self image of our most powerful AI (so far).
-
$500K Red Teaming Challenge Launches for Open Source Safety
By
–
We’re launching a $500K Red Teaming Challenge to strengthen open source safety. Researchers, developers, and enthusiasts worldwide are invited to help uncover novel risks—judged by experts from OpenAI and other leading labs.
-
When Will AI Surpass Humans Enhanced by AI?
By
–
The question is not when will AI be better than humans, but when will AI be better than humans with AI.
-
Fearlessness Determines Success in AI Competition
By
–
In AI as in everything, the fearless will win and the fearful will lose.
-
OpenAI Reveals AI Reasoning Chains Without Supervision Requirements
By
–
OpenAI se quita el pudor de enseñar sus Cadenas de Razonamiento y dejará que estos modelos open source las muestre sin ningún tipo de supervisión. Es gracioso porque como indican, una cosa es lo que la IA muestre como resultado y otra "lo que piense". Quizás viendo su
-
GPT-OSS Models Implement Advanced Safety Training Approaches
By
–
The gpt-oss models leverage our state-of-the-art approaches to safety training. We used deliberative alignment and an instruction hierarchy during post-training to help the models refuse unsafe prompts and defend against prompt injections, as well as pre-training interventions.