Why rigorous AI testing matters | Stephen Casper at FAR.AI's alignment workshop
Why rigorous AI testing matters | Stephen Casper at FAR.AI's alignment workshop
Видео Why rigorous AI testing matters | Stephen Casper at FAR.AI's alignment workshop канала FAR․AI
Видео Why rigorous AI testing matters | Stephen Casper at FAR.AI's alignment workshop канала FAR․AI
Комментарии отсутствуют
Информация о видео
17 апреля 2025 г. 21:00:02
00:00:56
Другие видео канала
![Kwan Yee Ng – AI Policy in China [Alignment Workshop]](https://i.ytimg.com/vi/JnU6QyjTJNE/default.jpg)
Kwan Yee Ng – AI Policy in China [Alignment Workshop]
Christian Schroeder de Witt - Secret Collusion Among GenAI Agents: a Model Evaluation Framework
Julian Stastny – Plan B: Training LLMs to fail less severely
Sophie Bridgers – Scalable Oversight: A Rater Assist Approach
Evan Hubinger – Deceptive Instrumental Alignment![Buck Shlegeris - AI Control and Why AI Security People Should Care [AI Security Forum]](https://i.ytimg.com/vi/tSj0GJVyROE/default.jpg)
Buck Shlegeris - AI Control and Why AI Security People Should Care [AI Security Forum]
Brad Knox - Your RLHF fine-tuning is secretly applying a regret preference model![Alex Wei – Paradigms & Robustness [Alignment Workshop]](https://i.ytimg.com/vi/0D6y7g7DL88/default.jpg)
Alex Wei – Paradigms & Robustness [Alignment Workshop]
Roger Grosse - Studying LLM Generalization through Influence Functions![Evan Hubinger – Alignment Stress-Testing at Anthropic [Alignment Workshop]](https://i.ytimg.com/vi/JfDlbzF6rsY/default.jpg)
Evan Hubinger – Alignment Stress-Testing at Anthropic [Alignment Workshop]![Atticus Geiger - State of Interpretability & Ideas for Scaling Up [Alignment Workshop]](https://i.ytimg.com/vi/eqZ1iEoor5s/default.jpg)
Atticus Geiger - State of Interpretability & Ideas for Scaling Up [Alignment Workshop]![Mahmoud Ghanem - Cyber Range Design at UK AISI [AI Security Forum]](https://i.ytimg.com/vi/okikt2qIqR4/default.jpg)
Mahmoud Ghanem - Cyber Range Design at UK AISI [AI Security Forum]![Joel Leibo – AGI-Complete Evaluation [Alignment Workshop]](https://i.ytimg.com/vi/Py_Wk2b2rvY/default.jpg)
Joel Leibo – AGI-Complete Evaluation [Alignment Workshop]
Paul Riechers & Adam Shai – Simplex![Richard Ngo – Reframing AGI Threat Models [Alignment Workshop]](https://i.ytimg.com/vi/4v3uqWeVmco/default.jpg)
Richard Ngo – Reframing AGI Threat Models [Alignment Workshop]
FAR.Research: Planning in a recurrent neural network that plays Sokoban
Vienna Alignment Workshop Panel Discussion – Current Issues in AI Safety
David Krueger - Sociotechnical AI safety
FAR Research: Adversarial Policies Beat Superhuman Go AIs![Adam Gleave – Will Scaling Solve Robustness? [Alignment Workshop]](https://i.ytimg.com/vi/kqDEkzSbwwo/default.jpg)
Adam Gleave – Will Scaling Solve Robustness? [Alignment Workshop]
Boaz Barak - The impossibility of (strong) watermarking