Community feed
A focused stream of recent stories from the sources curated for this community. Latest: Early Results on Monitorability in QA Settings, Introducing RTEB: A New Standard for Retrieval Evaluation, and Improving support with every interaction at OpenAI. Page 13.
Vincent Cheng, Thomas Kwa, and Neev Parikh share research on how AI agents can hide secondary task-solving from monitors, finding that harder tasks are more detectable and small models can learn to evade larger monitors.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Learn how OpenAI uses AI to enhance support, cutting response times, improving quality, and scaling to meet hypergrowth.
Google Stax Aims to Make AI Model Evaluation Accessible for Developers infoq.com
OpenAI introduces GDPval, a new evaluation that measures model performance on real-world economically valuable tasks across 44 occupations.
Cambridge scientists’ Trismik snaps £2.2M to redefine AI model evaluation using psychometrics Tech Funding News
IBM named a leader in the 2025 IDC Marketscape Worldwide GenAI Model Evaluation IBM
MITRE and FAA Introduce Novel Aerospace Large Language Model Evaluation Benchmark The MITRE Corporation
Apollo Research and OpenAI developed evaluations for hidden misalignment (“scheming”) and found behaviors consistent with scheming in controlled tests across frontier models. The team shared concrete examples and stress tests of an early method to reduce...
Intel Chips Excel in AI Benchmark: Will it Boost Prospects? Zacks Investment Research
NAVER D2SF Invests in Podonos, a Voice AI Model Evaluation Startup Based in North America PR Newswire
OpenAI’s new research explains why language models hallucinate. The findings show how improved evaluations can enhance AI reliability, honesty, and safety.
More stories load automatically as you scroll.