Fresh daily
AI News
Latest AI tool releases, research breakthroughs, and industry news.
Older
Democratic inputs to AI grant program: lessons learned and implementation plans
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.
Building agricultural database for farmers
Digital Green uses OpenAI to increase farmer income.
Increasing accuracy of pediatric visit notes
Summer Health reimagines pediatric doctor’s visits with OpenAI.
Weak-to-strong generalization
We present a new research direction for superalignment, together with promising initial results: can we leverage the generalization properties of deep learning to control strong models with weak supervisors?
AI-Exploits: Repo of multiple unauthenticated RCEs in AI tools
Article URL: https://github.com/protectai/ai-exploits Comments URL: https://news.ycombinator.com/item?id=38291880 Points: 67 # Comments: 18
Frontier risk and preparedness
To support the safety of highly-capable AI systems, we are developing our approach to catastrophic risk preparedness, including building a Preparedness team and launching a challenge.
DALL·E 3 system card
GPT-4V(ision) system card
Using GPT-4 for content moderation
We use GPT-4 for content policy development and content moderation decisions, enabling more consistent labeling, a faster feedback loop for policy refinement, and less involvement from human moderators.
Confidence-Building Measures for Artificial Intelligence: Workshop proceedings
Frontier Model Forum
We’re forming a new industry body to promote the safe and responsible development of frontier AI systems: advancing AI safety research, identifying best practices and standards, and facilitating information sharing among policymakers and industry.
Accurately analyzing large scale qualitative data
Viable uses GPT-4 to analyze qualitative data at a revolutionary scale with unparalleled accuracy.
Improving mathematical reasoning with process supervision
We've trained a model to achieve a new state-of-the-art in mathematical problem solving by rewarding each correct step of reasoning (“process supervision”) instead of simply rewarding the correct final answer (“outcome supervision”). In addition to boosting performance relative to outcome supervision, process supervision also has an important alignment benefit: it directly trains the model to produce a chain-of-thought that is endorsed by humans.
Language models can explain neurons in language models
We use GPT-4 to automatically write explanations for the behavior of neurons in large language models and to score those explanations. We release a dataset of these (imperfect) explanations and scores for every neuron in GPT-2.
GPTs are GPTs: An early look at the labor market impact potential of large language models
Preserving languages for the future
How Iceland is using GPT-4 to preserve its language.
Powering virtual education for the classroom
Khan Academy explores the potential for GPT-4 in a limited pilot program.
Forecasting potential misuses of language models for disinformation campaigns and how to reduce risk
OpenAI researchers collaborated with Georgetown University’s Center for Security and Emerging Technology and the Stanford Internet Observatory to investigate how large language models might be misused for disinformation purposes. The collaboration included an October 2021 workshop bringing together 30 disinformation researchers, machine learning experts, and policy analysts, and culminated in a co-authored report building on more than a year of research. This report outlines the threats that language models pose to the information environment if used to augment disinformation campaigns and introduces a framework for analyzing potential mitigations. Read the full report here.
Creating next-gen characters
Using GPT-3 to create the next generation of AI-powered characters.
The power of continuous learning
Lilian Weng works on Applied AI Research at OpenAI.