Confidence-Building Measures for Artificial Intelligence: Workshop proceedings

We have partnered with international news organizations Le Monde and Prisa Media to bring French and Spanish news content to ChatGPT. Post Views: 33

We’re developing a blueprint for evaluating the risk that a large language model (LLM) could aid someone in creating a biological threat. In an evaluation involving both biology experts and students, we found that GPT-4 provides at most a mild uplift in biological threat creation accuracy. While this uplift is not large enough to be conclusive,…

The results of today’s neural networks in fields as diverse as language, mathematics, and vision are remarkable. These networks, however, typically employ elaborate structures that are resource-intensive to run. When dealing with limited resources, such as those found in wearables and smartphones, delivering such models to users can be impracticable. Pruning pre-trained networks entails deleting…

We explore large-scale training of generative models on video data. Specifically, we train text-conditional diffusion models jointly on videos and images of variable durations, resolutions and aspect ratios. We leverage a transformer architecture that operates on spacetime patches of video and image latent codes. Our largest model, Sora, is capable of generating a minute of…

Working together to create open-source and private datasets for AI training. Post Views: 41

We funded 10 teams from around the world to design ideas and tools to collectively govern AI. We summarize the innovations, outline our learnings, and call for researchers and engineers to join us as we continue this work. Post Views: 43