Model Evaluation and Threat Research
4 open roles
description
Model Evaluation and Threat Research is a nonprofit focused on evaluating the capabilities and alignment of advanced machine learning models. They have been working with leading AI labs such as Anthropic and OpenAI to perform preliminary evaluations of their models, as practice building relevant institutional capacity and developing evaluation methodology.
Open roles
Learn more
80,000 Hours links
AI safety technical research
Career review
Long-term AI policy strategy research and implementation
Career review
External content
Beth Barnes, METR project lead, on safety evaluations and standards for AI
Talk
An overview of the goals of METR's team
Approach
Evaluating Language-Model Agents on Realistic Autonomous Tasks
Report
Collection of METR's resources for evaluating potentially dangerous autonomy capabilities of AI models
Resource
METR's 2023 Year in Review
Blog post
Common Elements of Frontier AI Safety Policies
Blog post