arXiv AI
Jul 9
Measuring the metacognition of AI
arXiv:2603. 29693v3 Announce Type: replace Abstract: A robust decision-making process must take into account uncertainty, especially when the choice involves inherent risks.
Confidence-Building Measures for Artificial Intelligence: Workshop proceedings
Related stories
arXiv AI
Sep 2
DualStake: Dual-Path Confidence Calibration in Deep Research Agents
DualStake introduces a dual-path confidence calibration for deep research agents, adding step confidence elicitation after each retrieval step. The method shows that evidence confidence (E-Conf) after the final retrieval provides a stronger uncertainty signal than answer confidence (A-Conf), and that A-Conf is largely influenced by E-Conf. By applying margin‑clipped, confidence‑dependent stake rewards, DualStake aligns both E-Conf and A-Conf with answer correctness, improving calibration across multiple QA benchmarks without harming accuracy.
arXiv AI
Aug 13
On Benchmarking Human-Like Intelligence in Machines
arXiv:2502. 20502v2 Announce Type: replace Abstract: Recent advances in Artificial Intelligence (AI) have yielded powerful computational models that, by learning from vast amounts of human-generated data, are increasingly posited as approximate models of human cognition.
arXiv AI
Jun 6
A Framework for Measuring Appropriate Reliance on Set-Valued AI Advice
arXiv:2606. 06081v1 Announce Type: new Abstract: Appropriate reliance on AI advice has become a central research theme in human-AI collaboration.
arXiv AI
Aug 25
The Measurement Revolution? Credible Measurement and Inference in the Age of AI
The article discusses how artificial intelligence is reshaping measurement in economics by converting unstructured data into structured variables at low cost, enabling large‑scale measurement that was previously infeasible. It outlines three stages—discovery, construct definition, and observation—where AI impacts the measurement pipeline and stresses the importance of rigorous validation to ensure credible inference. The review offers guidance on navigating the shift from a single scalable measure to multiple plausible ones that can lead to differing empirical conclusions.
arXiv Machine Learning
Jul 9
Robust Human-AI Complementarity under Uncertainty
arXiv:2607. 06656v1 Announce Type: new Abstract: Machine learning models are often intended to augment rather than replace human decision makers, by providing information that is complementary to human judgement.
OpenAI Blog
Jun 21, 2016
Concrete AI safety problems
We (along with researchers from Berkeley and Stanford) are co-authors on today’s paper led by Google Brain researchers, Concrete Problems in AI Safety. The paper explores many research problems around ensuring that modern machine learning systems operate as intended.
arXiv AI
Jun 3
Towards a Science of AI Agent Reliability
arXiv:2602. 16666v3 Announce Type: replace Abstract: AI agents are increasingly deployed to execute important tasks.
arXiv AI
Jul 10
Psychological Competence as a Missing Dimension in AI Evaluation
arXiv:2607. 08285v1 Announce Type: new Abstract: Current AI evaluation frameworks focus primarily on technical performance, including accuracy, robustness, reasoning ability, and policy compliance.
OpenAI Blog
Nov 6, 2025
AI progress and recommendations
AI is advancing fast. We have the chance to shape its progress—toward discovery, safety, and a better future for everyone.
OpenAI Blog
Oct 26, 2023
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.
arXiv AI
Jul 1
Wisdom Of The (AI) Crowd: Investigating Artificial Swarm Intelligence In Large Language Models
arXiv:2606. 31404v1 Announce Type: new Abstract: Human swarm intelligence demonstrates remarkable collective accuracy but faces scalability constraints in cost, coordination, and time.